117 Commits

Author	SHA1	Message	Date
Tai An	a7fa678d83	fix(tts): forward the OpenAI speed field to the backend (#11097) (#11120) ... * fix(tts): forward the OpenAI speed field to the backend (#11097) /v1/audio/speech accepted the documented OpenAI `speed` field and then dropped it: schema.TTSRequest had no Speed member, so the value never reached proto.TTSRequest and the request returned 200 with an unchanged playback rate. Accept speed and normalise it into the existing per-request params map, which core/backend forwards verbatim to the backend. An explicit params["speed"] still wins, and a value outside the documented 0.25-4.0 range is now rejected with 400 instead of being silently ignored. Signed-off-by: Anai-Guo <antai12232931@outlook.com> * fix(tts): distinguish explicit speed=0 from an omitted field Make TTSRequest.Speed a *float32 so an explicit `"speed": 0` (invalid, below the documented 0.25 minimum) is rejected with 400 instead of being treated as unset and silently defaulted. An omitted field stays nil and leaves the backend default untouched. Add a request-boundary regression that distinguishes an omitted speed from an explicit zero, addressing review feedback. Signed-off-by: Anai-Guo <antai12232931@outlook.com> * docs: drop the speed field from the TTS docs Per review: no backend consumes params.speed today, so documenting it would be misleading. The API-level plumbing and validation stay. Signed-off-by: Anai-Guo <antai12232931@outlook.com> --------- Signed-off-by: Anai-Guo <antai12232931@outlook.com>	2026-07-27 19:03:32 +02:00
mudler's LocalAI [bot]	83a0f16a21	feat(gallery): let one gallery entry offer several builds of the same model (#10943) ... * feat(system): expose raw detected capability for model meta resolution Model meta gallery entries express hardware fallback through candidate ordering rather than a capability map, so they need the undecorated detected capability string without Capability's default/cpu fallback chain. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactor(system): drop duplicate capability accessor, cover DetectedCapability ReportedCapability was added with a body identical to the existing DetectedCapability. Keep one accessor and move the specs onto it, since DetectedCapability had no direct coverage of its no-fallback behavior. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(vram): parse IEC binary size suffixes (KiB..PiB) ParseSizeString accepted only SI suffixes, so a "20GiB" floor was rejected outright. Model and VRAM sizes are conventionally quoted in IEC units, and silently reading GiB as GB would understate a floor by about 7%. Purely additive: these inputs previously returned an unknown-suffix error. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): add Candidate type for meta model entries Candidate is one option in a meta entry's ordered variant list. It names a concrete gallery entry and declares when that entry suits the host. EffectiveMinVRAM resolves the VRAM floor, letting an authored min_vram win over a nightly-inferred one. An unparseable floor errors instead of being treated as absent: swallowing a typo would turn a constrained candidate into an unconstrained one and select a too-large variant rather than fail loudly. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): add hardware-aware model variant resolver Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): allow gallery model entries to declare variant candidates A gallery entry with a non-empty candidates list is a meta entry: it names an ordered list of concrete entries and resolves to the first one the host can satisfy, instead of describing model files directly. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): resolve meta model entries to hardware-appropriate variants at install Meta gallery entries carry an ordered candidate list; at install time the first candidate the host satisfies is resolved and its payload installed under the meta's name, so the model keeps a stable name regardless of which variant backs it. The resolution is recorded in the installed gallery config so a reinstall honors a prior pin and operators can see the backing variant. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): key meta pin recall on the installed name and detach resolved entries Six review findings on the meta-entry install path. Pin recall was keyed on the gallery entry name while applyModel writes the record under the install name (req.Name when supplied), so a meta installed under a custom name with a pin lost that pin on reinstall and was silently re-resolved onto a different variant, possibly swapping its backend. Compute the install name with applyModel's own precedence before the recall. ResolveMetaModel returned a shallow struct copy, so the resolved entry's Overrides aliased the gallery entry's map and the install path's in-place mergo merge wrote the caller's request into the shared catalog. Detach Overrides, ConfigFile, AdditionalFiles, URLs and Tags. Not exploitable today only because this path re-unmarshals the gallery per call, which is a property nobody should have to rely on. Also: overlay the meta's name onto the persisted config for meta installs so the gallery file no longer records the variant's name; move the pinned-VRAM warning below the variant validation so a pin naming a nonexistent entry does not warn about VRAM before failing for an unrelated reason; and stop seeding config.URLs in the config_file branch, which duplicated every declared URL. Add seven network-free specs driving InstallModelFromGallery with a meta entry: variant payload wins over the meta's legacy url fallback, the resolution record round-trips to disk, a pin is recorded and honored on reinstall including under a custom install name, and the resolved entry does not alias the gallery's maps. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): deep-copy meta overrides and make two specs functional ResolveMetaModel detached the resolved entry's Overrides and ConfigFile with maps.Clone, which only copies the top level. Gallery overrides are nested in practice (parameters.model is near-universal) and the install path merges the caller's request with mergo.WithOverride, which recurses into nested maps and overwrites them in place, so the gallery entry's own inner maps were still reachable and still got rewritten by the last caller to install. Copy both maps all the way down instead, recursing through the container shapes a YAML decoder produces. ConfigFile is not mutated on the install path today, but it carries the same kind of nested payload and leaving it shallowly cloned would invite the bug back. Also fix two specs that passed whether or not their target fix was present: - "does not write the caller's overrides back into the gallery entry" re-read the catalog from disk, which re-unmarshals fresh structs and so cannot observe in-memory aliasing. It now asserts against the in-memory gallery entry and drives the real mergo merge. - "round-trips the resolution record to disk under the meta's name" asserted a name that is already correct in the config_file branch. It now drives the url branch via a file:// fixture, where the meta-name overlay actually applies. Both were verified red by reverting their fix. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * test(gallery): lint meta model entry invariants in index.yaml Adds Ginkgo specs that parse the shipped gallery/index.yaml and enforce the invariants that keep meta entries safe: a legacy url fallback equal to the final candidate's url, references only to existing non-meta entries, a min_vram floor on every candidate but the last-resort one, a capability drawn only from the vocabulary the system can report, and descending VRAM floors within a capability group. The capability check is the only compensating control for a typo there. Candidate matching is a case-sensitive exact comparison against SystemState.DetectedCapability(), so an unknown value never matches and falls through silently instead of erroring. The vocabulary therefore mirrors the raw return set of getSystemCapabilities(), which notably excludes "cpu": that is a fallback key inside Capability(capMap) on the meta backend path, never a reported capability. A CPU-only host reports "default". These pass vacuously until the pilot meta entry lands; the guard is intentionally in place before the thing it guards. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * test(gallery): close coverage gaps in the meta entry lint The ordering invariant grouped candidates by capability and asserted floors descend within a group. A candidate with an EMPTY capability matches every host, so it does not belong in its own group: it dominates every later candidate whose floor is at or above its own, across capability groups. Track a running minimum floor over the unconditional candidates instead, which subsumes the old same-group check for the empty capability. Every spec skipped non-meta entries, so with zero meta entries in the index all five bodies were no-ops. Aligning GalleryModel.IsMeta() with GalleryBackend.IsMeta(), whose semantics are deliberately opposite, would have made all of them pass while checking nothing. Extract each invariant into a helper over a slice of entries returning the violations it finds, and cover those helpers with synthetic fixtures so the logic stays tested at zero meta entries. The index-driven specs are now a thin application of already proven logic. Also assert the index parses non-empty, report every violation in one run rather than aborting on the first, and parse the index once for the suite. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci(gallery): add nightly denormalization of meta model candidates Fills the read-only backend, quantization and inferred_min_vram fields on meta gallery candidates and opens a PR, modeled on the existing checksum_checker job. Computing these needs network access, so it happens nightly rather than at install time. An authored min_vram is never modified: a human who measured a real load knows more than a pre-download estimate does. The index is rewritten via yaml.Node rather than a document round-trip. A full round-trip reflows all ~26k lines of gallery/index.yaml, which would bury the computed values and make the nightly PR unreviewable. The rewrite touches only the three derived keys, so authored styling survives and a run that computes nothing leaves the file untouched. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(ci): keep the gallery denormalize diff reviewable and self-healing The nightly denormalization job edits YAML nodes instead of round-tripping structs so its PR stays small enough for a human to review, but the write path undid that: yaml.Marshal re-encoded the node tree at yaml.v3's default 4-space indent and dropped the leading document marker, reflowing roughly 6000 lines around the handful of real changes. Encode through yaml.NewEncoder at the index's authored 2-space indent and restore the header. A write that changes three fields now changes three lines. Stale inferred_min_vram values were also never cleared. Both skip paths (an authored min_vram is present, or the candidate is the last resort) returned before touching the field, so a candidate that gained a floor or became the last resort after a reorder kept an inferred value that EffectiveMinVRAM reported as a real constraint, failing the meta lint with no way for the job to self-heal. Clear the field before both skips. The workflow discarded a whole night's work on any single failure: the program exits 1 when a candidate cannot be estimated, which aborted the job before the PR step, so one unreachable candidate blocked every other refresh indefinitely. Capture the status, open the PR with what was computed, mark the PR body as partial, and fail the run afterwards so the problem still surfaces. Also preserve the index's existing file mode instead of forcing 0644, and drop the redundant //go:build ignore tag, since Go already skips dot directories and the sibling modelslist.go carries no tag. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): add nanbeige4.1-3b meta entry with hardware-resolved variants Adds the first real meta entry to the gallery index. It resolves to the Q8_0 build on hosts with at least 6GiB of VRAM and to the Q4_K_M build everywhere else, installing either payload under the stable name nanbeige4.1-3b. The entry carries a url equal to its final candidate's url. LocalAI releases that predate candidates support parse the index non-strictly and drop the key silently, so without that url they would list the entry and install nothing. A regression spec parses the index the way those releases do and asserts every meta entry stays installable for them. Also teaches core/schema/gallery-model.schema.json about candidates. The schema sets additionalProperties: false at the top level, so an author following CONTRIBUTING.md and adding the yaml-language-server comment would otherwise get a validation error on this entry. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): make candidate entries complete, installable entries Reworks hardware-resolved gallery variants after a design pivot. There is no longer a separate "meta" entry kind. A gallery entry is a normal, complete entry that may additionally carry candidates:, a list of hardware-gated upgrades over itself, and the entry is itself the last-resort candidate. The previous design relied on a bare url: as the fallback for LocalAI releases that predate candidates support. That fallback is empty in practice: none of the 80 gallery/*.yaml files carry a top-level files:, and 1216 of 1281 index entries carry their payload in the index entry itself, so a url alone yields a config template with nothing to download. Since every released LocalAI reads gallery/index.yaml live from master, merging a payload-less entry would have shown every existing user a model that installs to a broken state. Making the entry its own base candidate removes the problem at the root: old clients drop the candidates key and install the entry exactly as they do today. Resolution order is now explicit pin, then capability plus VRAM over the declared upgrades, then the entry itself. The entry ALWAYS installs: when its own min_vram or capability is unmet the installer warns and installs it anyway, because there is nothing below it and refusing would make the gallery behave worse the newer the client is. A pin naming the entry's own name is valid and is how an operator declines an upgrade. IsMeta() becomes HasCandidates(), ResolveMetaModel becomes ResolveVariant, and the persisted meta_name record key becomes entry_name. GalleryBackend.IsMeta() is a separate concept and is untouched. The lint drops the three rules the pivot makes wrong (url equality with the final candidate, no inline payload, unconstrained final candidate) and gains one: the entry's own floor must sit strictly below every candidate's, since a base that outranks a candidate makes that candidate unreachable. The pilot entry is now the existing nanbeige4.1-3b-q4, which gains a 2GiB floor of its own and a single 6GiB upgrade to nanbeige4.1-3b-q8, replacing the separate nanbeige4.1-3b entry added in d0d441bb4. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): select model variants by hardware fit, not authored order Gallery entries could already carry a list of alternatives, but selection was an authored, ordered, first-match policy: every candidate declared a `capability` string and the VRAM floors had to descend in a hand-tuned order. That pushed hardware knowledge onto whoever edits the gallery and made ordering load-bearing, so a reordered list silently changed what users installed. None of it was necessary. SystemState.IsBackendCompatible already derives hardware support from a backend name alone: it knows MLX and metal are Darwin-only, CUDA is NVIDIA-only, ROCm AMD-only, SYCL Intel-only. Selection can read that instead of asking authors to restate it. Authoring is now just a list of names: - name: qwen3.6-27b min_memory: 4GiB variants: - model: qwen3.6-27b-mlx-8bit - model: qwen3.6-27b-gguf-q8 min_memory: 28GiB and all the intelligence moved into the selector. Given a host it drops the variants whose backend cannot run here, drops those whose known memory requirement exceeds what the host has, and takes the LARGEST of what is left, because a bigger footprint is a higher quality quantization of the same model. A variant of unknown size is kept, since nothing proves it does not fit, but it ranks last so a proven fit always beats a guess. An explicit pin still wins outright, and if nothing survives the entry installs its own payload: the base always installs, this never refuses. Available memory is VRAM when a GPU was detected and system RAM otherwise, read through xsysinfo so a cgroup limit is honored and a container gets its own limit rather than the node's RAM. Capability disappears entirely, from the types, the schema and the lint. VRAM and RAM collapse into one `min_memory`, because a model's footprint is roughly the same wherever it lives and one figure is compared against whichever applies. The lint rules about ordering, the capability vocabulary and floor relationships are deleted with the hazards they described; what remains is that every variant names an entry that exists and does not itself declare variants, plus that any memory figure actually parses. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * gallery: size model variants with a live probe, drop the nightly denormalizer Selection needs each variant's size to decide whether it fits and to rank largest-first. That figure was written into the index by a nightly job, which made the gallery carry a derived value that could drift from the entry it was derived from. Derive it at install time instead. pkg/vram already sizes a model without downloading it, and the gallery UI already uses it: a remote GGUF header range-fetch, then an HTTP HEAD for the content length, then any declared size:. It caches its results, so reuse it rather than writing a second probing path. A probe failure must never fail an install, so an unprobeable variant is treated as unknown: it survives the memory filter, because nothing proves it does not fit, and it ranks last, so a known-good fit always beats a guess. If every probe fails, selection still terminates on the base entry. The probe is injected through ResolveEnv rather than called directly, for the same reason the backend compatibility check is: specs pin an exact size, or an exact failure, without reaching the network. With that in place three things are dead weight and go: - The nightly job and the fields it populated. Variant.Backend was redundant because the backend is resolved live from the referenced entry during selection, and Quantization was display-only that nothing read. - min_memory on the base entry. The base always installs and its floor could only warn, so it could not change any outcome. - The lint rules and schema entries for both. min_memory on individual variants stays, as the override for when the probed size is wrong. An authored figure now suppresses the probe entirely rather than merely outranking it, so it costs no round trip. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): expose model variants for selection over API, CLI and MCP A gallery entry may carry `variants:`, alternative builds of the same model. Selection already worked at install time, but nothing could see what an entry offered or ask for a specific build, so the feature was undrivable. Listing: `GET /api/models` now reports `variants` and `auto_variant` for the entries that declare variants. Each variant carries its resolved backend, its measured size and whether it fits this host. `auto_variant` is what installing without a choice would pick right now. The new gallery.DescribeVariants runs the same variantOptions + SelectVariant pass the installer runs, so the reported default cannot drift from what installing actually does, and HostResolveEnv is extracted so both derive the host and share pkg/vram's probe cache from one place. Performance: an entry that declares no variants returns early without touching the probe, so the ~1280 ordinary entries cost exactly what they cost before. Selection: `variant` is accepted on POST /models/apply, as a query param on POST /api/models/install/:id, on the gallery apply file/string request, as `local-ai models install --variant`, and as a parameter on the install_model MCP tool (both the httpapi and inproc clients). Empty means auto-select. An unknown variant name now fails the install naming what was requested. This closes a real hole: an entry declaring no variants short-circuits before selection runs, so a requested variant was previously dropped silently and the install reported success. startup.InstallModels ends in a variadic model list, so install options could not be appended to it; InstallModelsWithOptions is added alongside and InstallModels delegates to it. No caller signature changed. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): drop the redundant variant min_memory field Variant.MinMemory was an authored override for when the live probe misreads a variant's footprint. It duplicated an existing field: probeEntryMemory already passes the entry's declared size: into EstimateModelMultiContext, whose cascade prefers that declared size over its own guesswork. Correcting size: on the referenced entry fixes the figure for every consumer rather than only for variant selection, so min_memory shadowed the right answer. A variant is now nothing but a name. Its effective size is exactly the probe result, and an unknown stays unknown: it survives the filter and ranks last. EffectiveMemory loses its error return along with the field. The authored string was the only thing that could fail to parse, so the error had no remaining source and was propagating dead nil-checks through SelectVariant, DescribeVariants and the pin warning. Selection behaviour is unchanged. The specs covering probe-derived sizing, ranking, filtering, the unknown-size path, pin recall, entry/variant metadata split and deep-copy isolation all survive; the three install specs that needed a definite size now declare it through the referenced entry's own size:, which exercises the documented escape hatch directly. gallery/index.yaml is untouched: no entry ever carried the key. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): rank the entry's own build against its variants Variant selection pulled the declaring entry's own payload, the base, out of the candidate set and consulted it only once every declared variant had been rejected. Two real failures followed. A variant whose size the probe cannot determine deliberately survives the memory filter, because nothing proves it does not fit. As the only survivor it then won outright on any host, however small: a 2GiB machine installed an unmeasured variant in preference to the 4GiB build the entry itself ships, with no warning. 241 of the 1280 current index entries carry no files and no size, which is exactly that shape. "Largest wins" also broke whenever the base was the largest. An author writing a Q8 entry that offers a Q4 downgrade for small hosts, a natural shape that nothing in the lint, schema or docs discourages, had the Q4 installed on every large host instead. Make the base an ordinary participant. It is still exempt from both filters, so selection always terminates on something installable, but it is now ranked against the variants: a proven fit first and largest, then the base, then any variant whose size nothing could measure. Both failures disappear together. The base is probed for its size accordingly, which it was not before, because an unsized base would lose every contest to an unmeasurable variant. FellBackToBase is kept but narrowed to "no declared variant survived", rather than "the base was chosen", since the base now also wins on merit and that is not worth warning about. A recalled variant pin also became a permanent install failure. A pin the caller supplies on this request must stay fatal, but one recalled from ._gallery_<name>.yaml can be invalidated by any later gallery edit, and failing on it turned one rename into a model that could never be reinstalled or upgraded again short of deleting a dotfile the user has never heard of. A stale recalled pin is now dropped with a warning naming it, and selection runs as if it had never been recorded. Also drop the last textual reference to two abandoned designs from the DetectedCapability comment, correct the documented variants JSON example, which showed a memory_bytes of 0 that omitempty makes impossible, and remove an em dash from the install skill. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): budget variant memory from RAM when a GPU reports no VRAM Variant selection read its memory budget from VRAM whenever a GPU capability was detected, and from system RAM only when none was. Apple Silicon satisfies the first branch and fails the premise: arm64 macs report the metal capability unconditionally, without probing anything, while TotalAvailableVRAM has no discrete VRAM pool to find and returns zero. The budget therefore came out as zero on every Mac. Zero drops every variant carrying a known size, so the base build was installed on all of them however much memory the machine had. The feature was inert on the platform, and silently: falling back to the base is a legitimate outcome, so nothing looked wrong. Take VRAM only when it is actually a number, and fall back to RAM otherwise. On a unified-memory host RAM is not an approximation of the budget, it is the budget, since the GPU shares it. A discrete GPU whose VRAM could not be read also lands on RAM, which overstates what the card holds but understates nothing the host has; the previous zero understated both. An unreadable RAM figure still yields zero and still installs the base, so a genuinely unknown host is not talked into a larger download. This is what turned tests-apple red: "installs a fitting variant's payload under the entry's own name" asserts on selection, and the runner resolved to the base because its budget was zero. The added specs pin the branch directly rather than relying on a macOS runner to notice again. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): add a model variant picker to the models gallery PR #10943 shipped the server side: a gallery entry may declare `variants:`, `GET /api/models` attaches `variants` and `auto_variant` to declaring entries, and `POST /api/models/install/:id` accepts a `variant` query parameter. Nothing in the UI consumed any of it, so the feature was not reachable from the browser. This wires it up. modelsApi.install takes an optional second argument and appends an encoded `?variant=` only when one is given, so every existing call site keeps sending exactly the request it sent before. On the models table, an entry that declares variants gets a split button. The primary Install still installs the auto-selected build, because auto is the default and the point of the feature; the chevron opens a menu for a deliberate override. It follows the Backends.jsx precedent: one shared Popover re-anchored per row, rendering .action-menu items, which brings Escape, outside-click and focus return along with it. An entry that declares no variants renders exactly as it did before. A variant that does not fit is dimmed but stays selectable, since the server honors an explicit choice with a warning rather than refusing it. memory_bytes is omitempty on the wire, so an absent key means the size is unknown and never zero. A single helper guards both the menu and the detail row, because formatBytes would otherwise render a falsy value as "0 B", which reads as "needs nothing". The expanded detail row gains a Variants section listing each build's backend, size, whether it fits, which is the entry's own build, and which one auto-selection would pick, built from the existing DetailRow helper and .badge classes. Eight Playwright specs cover the picker, including that plain Install sends no variant parameter and that choosing one sends it. One pre-existing assertion was scoped with .first(): the Variants section legitimately adds more llama-cpp badges to the detail row, which tripped strict mode. UI line coverage 49.42% -> 49.36% against a 40.0 baseline and 0.8pp tolerance; branch coverage rose 72.04% -> 72.66%. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): describe model variants from a companion endpoint Variant description probes each referenced entry's weight files over the network: an HTTP HEAD plus a ranged GET, serial, five seconds per probe with no aggregate deadline. Running it inline in GET /api/models made one listing cost (entries x variants) round trips. The Manage page fetches with items=9999, so at 200 declaring entries that is ~1000 serial probes, minutes of a blocked handler and gigabytes of range traffic for a single page load. Only one entry declares variants today, but the feature exists so that many will. Follow the precedent already set for VRAM estimates. The listing now reports only has_variants, a length check on loaded metadata that touches nothing, and GET /api/models/variants/:id returns the description for one entry, mirroring estimate/:id in route shape, auth and error handling. DescribeVariants itself is unchanged; only its caller moved. The picker fetches lazily at the two points where a user asks to see variants, opening the split-button menu and expanding the detail row, and caches per entry for the page session. An entry declaring no variants issues no request at all. A spec counts real HTTP hits on the weight files, so it goes red if description becomes reachable from the listing path again through any caller. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): filter the model gallery to entries that declare variants The gallery is heading towards showing parent entries and hiding the individual builds they reference, so a user sees one row per model rather than six quantizations of it. Adoption is a single entry today, so defaulting to that would leave a one-row gallery. This ships the migration-phase inverse instead: the default is untouched, and a toggle narrows the list to only the entries that declare variants. It previews the end state and changes nothing until someone asks for it. The filter is server-side, next to term/tag/backend/capability and above the pagination arithmetic. The listing paginates at 9 items, so narrowing on the client would leave totalPages and availableModels describing the unfiltered set and hand the user empty pages. It selects on HasVariants(), which reads already-loaded metadata, so it issues no variant probes. The parameter is named has_variants after the listing field it selects on, and is compared against "true" like the other boolean query params (all_users, save_checkpoint), so has_variants=false reads as absent. With it omitted the response is byte-for-byte what it was before. The control is the shared Toggle component, matching the fitsFilter toggle already on this page: same wrapper class, same icon and label shape, same localStorage persistence. Unlike fitsFilter it resets to page 1 on change, which a server-side filter has to do. Stacking the toggle with a tag or backend filter easily yields nothing while one entry declares variants, so the empty state now names the variants filter as the cause rather than leaving a user to conclude the gallery is broken. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(ui): render gallery model descriptions as Markdown Gallery descriptions are Markdown, but the React UI dumped them raw, so a model whose description opens with an ATX heading showed a literal "# Qwen3.6-27B [](https://chat.qwen.ai)" in the list. Full-description areas now render through renderMarkdown (marked + DOMPurify), matching how Backends.jsx and the Manage detail panels already handle the same content: - Models.jsx expanded detail row - VoiceLibrary.jsx voice detail header The truncated one-line previews must not render block Markdown: a leading "#" would become an <h1> and wreck the row height and rhythm. They get a new stripMarkdown() helper instead, which reduces Markdown to a single line of readable plain text. It is used for the cell text and for the title tooltip, since a tooltip full of "[](url)" is no better than a cell full of it: - Models.jsx gallery table description cell - Manage.jsx model and backend resource-row descriptions stripMarkdown walks marked's lexer output rather than running regexes over the source, so what it strips is by construction what renderMarkdown would have rendered, and it needs no new dependency. Output lands in JSX text nodes, so React escapes it; no new dangerouslySetInnerHTML beyond the two full-description sites, both of which run DOMPurify. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(ui): strip Markdown from the backends table description cell Commit b35d630cf fixed this for gallery models but left the Backends admin page with the same asymmetry: its detail panel renders the description through renderMarkdown, while the collapsed table row dumped the raw gallery string into both the cell body and the title tooltip. That is user-visible. 40 of the 949 entries in backend/index.yaml carry Markdown - insightface uses inline code backticks, others use lists and links - and backend descriptions also contain embedded newlines, so the one-line cell showed literal syntax. The cell now runs stripMarkdown over the description once and uses the result for the text and the title, matching Models.jsx and the ResourceRowDesc component in Manage.jsx. The '-' placeholder is preserved, and now also fires when a description reduces to nothing after stripping. The detail panel is untouched and no new dangerouslySetInnerHTML is introduced: stripMarkdown output lands in a JSX text node, so React escapes it. Three Playwright specs cover it: a description with a heading, inline code and a link renders as clean text with no literal syntax and no block element in the cell, the title tooltip carries the same stripped text, and a backend without a description still shows the placeholder. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ui(models): polish the variant detail view and scope rendered Markdown The gallery detail pane rendered every field through the same two-column label/value row, including the description. Multi-paragraph prose in a value cell ran eight rows tall at the top of the pane on a ~1200px measure, breaking the grid's rhythm exactly where the eye enters. Move it into its own full-width block above the table, capped at a 68ch measure, keeping the label. Rendered Markdown had no scoped typography anywhere in the app, so a description opening with `#` inherited the browser default 2em inside a 13px surface while a `##` further down was indistinguishable from body text. Add a reusable .markdown-body block mapping h1-h6, paragraphs, lists, links, code, blockquotes, images and tables onto the existing type scale, and apply it to every renderMarkdown() consumer: the models detail, the backends detail, both Manage details and the voice library detail. Rebalance the variants list so the name leads. Backend and size drop from badge/secondary weight to muted metadata; the FITS badge goes entirely, since it was true of nearly every row and so said nothing, while the variant that does not fit keeps a warning badge and a dimmed name. AUTO-SELECTED stays marked because it answers what a plain Install produces. Rows share the parent's grid tracks via subgrid so name, backend, size and status line up down the list instead of raggedly following name length. Finally, make each variant row actionable. It looked like a list of choices but was inert text, with per-variant install hidden behind the split-button chevron elsewhere; each row is now a button onto the existing handleInstall(modelId, variant) path, with hover, keyboard focus and a disabled state while an install is in flight. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): collapse the listing to one row per model The listing supported has_variants=true, which narrowed to entries that DECLARE variants. With adoption at three entries that showed three rows, which is useless; it was always a placeholder. Replace it with the view that is actually useful: the deduplicated gallery. Show every entry installable in its own right and nothing twice, which means the parents plus every entry nobody references, and hide only the builds another entry already offers as a variant, since those are reachable through their parent. The parameter is renamed to collapse_variants accordingly: the filter is no longer a predicate on a row's own metadata but a view over the whole gallery. Default stays off, so the response with the parameter absent is unchanged. VariantReferencedIDs never reports an entry that declares variants of its own, so parents are always visible. That guarantees every hidden entry has a visible entry offering it, and no chain can strand a row. Variant resolution already refuses to install such a reference, but the listing has to stay coherent in the presence of a gallery that has one rather than silently swallowing entries. Self-references and dangling references hide nothing. The referenced set is computed over the whole gallery rather than over what the other filters left, so an entry is hidden because a parent offers it and never because of what the user searched for. The pass is over metadata already in memory: it resolves nothing over the network and triggers no variant description or size probe, so the listing's zero-probe contract still holds. The UI toggle keeps its behaviour (persistence, page reset, clear filters) and becomes "One row per model", which says what the user gets. Its localStorage key moves too, since the stored value meant a different filter. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): show the collapsed model listing by default The gallery listing is what a user reaches for to answer "what can I install". Answering that with several rows for the same model, one per build, makes the reader do the deduplication the collapsed view already does, so the collapsed view is the one to land on. The UI now asks for collapse_variants=true unless the toggle says otherwise. The server default is deliberately untouched: a request with the parameter absent still returns the full listing, because other API clients depend on that response and collapsing it under them would be a breaking change. Opting out omits the parameter rather than sending false, so it asks for exactly the listing everyone else gets. The stored preference changes vocabulary from '1'/'0' to 'on'/'off'. The previous build wrote it from an effect that runs on mount, so a stored '0' recorded that the page had been opened rather than that anyone chose the expanded view, and honouring it would pin every earlier visitor to a default they never picked. Only the new vocabulary counts as a choice; a legacy '1' meant the collapsed view and is what the new default gives anyway, so no earlier deliberate choice is lost. Collapsing being the default also changes what the empty state may say about it. An opted-into filter can be named as the cause of an empty result; a default cannot, so the filters keep the top line and the collapsed view drops to a hint below it, shown only once filters are narrowing the set. For the same reason "Clear filters" now restores the collapsed default instead of switching it off, and the toggle alone no longer counts as a filter worth offering to clear. The label stays "One row per model": it describes the view the user is looking at rather than an action, so it reads the same whether it is opted into or out of. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * gallery: group alternative builds of the same weights under variants Sweep the gallery for entries that are alternative builds of the same weights (different quantization, precision, or runtime format) and declare them as variants of a single parent row, so the listing offers one row per model instead of one row per quantization and the installer picks the largest build that this host can actually run. 41 families over 95 entries, turning 54 entries into variants. The parent is the bare-named entry wherever one exists, so nothing changes about what any existing entry installs. Ranking already selects the largest fitting build regardless of which entry is nominally the parent, so the parent only decides the pathological case where nothing fits. For the ten families that have no bare-named entry, the smallest build is the parent, since that is the one that has to install when nothing fits. Grouping was verified against the actual model filenames rather than the entry names alone. Different parameter sizes, languages, finetunes, and products that merely share a name prefix are left as separate rows: the qwen3.6 APEX and pi-tune finetunes, the DFlash and MTP speculative-decoding pairings, English-only versus multilingual Whisper, the QAT versus non-QAT Gemma 4 weights, and the abliterated FLUX build are all distinct models. Six parents define YAML anchors that other entries pull in with a merge key, which would have handed their variants to every merging child. For the two depth-anything anchors that would have made fourteen unrelated entries advertise the base model's builds as their own. All 26 merging children therefore carry an explicit empty variants list, which overrides the merged key and is equivalent to the key being absent. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): rank model variants by host backend preference Variant auto-selection filtered candidates by whether their backend can run on the host, then ranked the survivors by size alone. The backend never influenced the choice beyond that gate, so a Mac offered both an MLX build and a llama.cpp build kept neither filtered and installed whichever was larger, leaving the native accelerated runtime unused. The same held for CUDA against CPU on NVIDIA and ROCm against Vulkan on AMD. Rank by the host's backend preference between the fit tier and size: fit stays a filter, preference decides among the builds the host can equally hold, and size still separates builds on equally preferred runtimes. The preference data stays in one declarative table in pkg/system, now read by a prefix lookup instead of a switch, so adding a capability or reordering one host's runtimes is a one-line edit and the gallery's ranking code carries no per-backend branching. MLX joins the metal rule ahead of metal itself, which is inert for the existing alias-resolution consumer because no alias group holds a candidate named for mlx. An unrecognised backend, an unrecognised capability and an absent preference list all collapse to the previous size-only ordering rather than erroring or dropping candidates. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): rank variants by engine name, not backend build tag Variant auto-selection ranked candidates with SystemState.BackendPreferenceTokens, but that function and the variant ranker speak different vocabularies. BackendPreferenceTokens returns BUILD TAGS ("cuda", "rocm", "sycl", "vulkan", "metal", "cpu"). It exists to match installed backend build directory names like "llama-cpp-cuda-12" during alias resolution in ListSystemBackends. Variant ranking instead matches a gallery entry's `backend:` value, which is an ENGINE NAME: "llama-cpp", "vllm", "vllm-omni", "sglang", "mlx" and the rest. No engine name in gallery/index.yaml contains "cuda", "rocm", "sycl" or "vulkan". preferenceRank matches by substring, so on an NVIDIA host the tokens [cuda, vulkan, cpu] matched neither "llama-cpp" nor "vllm", every candidate scored identically and size alone decided. The NVIDIA, AMD, Intel, darwin-x86 and vulkan rules were all inert. Only metal appeared to work, and only because the token "mlx" happens to equal an engine name. The mismatch does not error, it silently deletes the feature. Separate the two vocabularies. backendBuildTagPreferenceRules keeps the build tags and its original output for every capability, including metal, whose "mlx" token is removed again; its alias-resolution consumer is byte-identical to before. engineNamePreferenceRules is new, holds engine names, and is read by the new EnginePreferenceTokens, which HostResolveEnv wires into the renamed ResolveEnv.EnginePreference. Both tables sit adjacent under one block comment naming each vocabulary and each consumer, and share one lookup helper so their semantics cannot drift. On NVIDIA the order is vLLM, then SGLang, then llama-cpp: vLLM is the throughput engine and a model published with a vLLM build is published that way because that build is the one worth running. AMD and Intel get the same order, since rocm and intel builds of both serving engines ship. Metal prefers mlx over llama-cpp. Vulkan prefers llama-cpp, the only LLM engine with a Vulkan build. darwin-x86 and unknown capabilities are deliberately absent rather than guessed at, degrading to the size-only ordering that predates preference. preferenceRank stays generic and names no engine and no capability, so adding a runtime remains a one-line table edit. Specs pin the NVIDIA and metal rules through the live table and the real HostResolveEnv wiring, so emptying the engine table or wiring the build tag source back in both go red. A regression table asserts BackendPreferenceTokens' original output per capability, and mirrored locks assert neither table carries the other's vocabulary. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * docs: record that variant selection ranks by engine before size A gallery entry can now declare variants, and selection ranks the builds a host can run by engine preference before size. Nothing told a contributor adding a backend that engineNamePreferenceRules exists, so a new engine would silently rank below every known one and lose to whatever build happened to be larger on hosts where it should have won. Document the step where a backend is added, warn against the sibling backendBuildTagPreferenceRules table (build tags, not engine names: the wrong table matches nothing, scores every candidate equally and disables the preference without erroring), and index it from AGENTS.md. Fix the authoring and user docs, which still claimed the largest surviving build wins. An author grouping builds under one entry has to be able to predict what a user gets, and size alone no longer decides it. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(cli,mcp): describe variant auto-selection as preference before size The CLI flag help and the install_model tool schema both still said auto-selection takes the largest build that runs. Ranking now puts engine preference ahead of size, so on NVIDIA a vLLM build wins over a larger llama.cpp one. An assistant reading the old schema would tell users the wrong thing. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): prefer llama.cpp over GPU serving engines on hosts with no GPU engineNamePreferenceRules had no row for the "default" capability, which getSystemCapabilities() returns both when no GPU is detected and when a GPU is present but under the 4 GiB VRAM floor. A missing row yields an empty preference list, which preferenceRank reads as "score everything equally", collapsing variant selection to size alone. That would be harmless if the hardware filter dropped GPU serving engines on such a host, but it does not. IsBackendCompatible derives support from the engine NAME, and "vllm" and "sglang" contain none of the darwin, cuda, rocm or sycl tokens it keys on, so they fall through to its closing "return true". A vLLM variant therefore survives on a CPU-only box and wins whenever its build is the larger of the two on offer: the machine installs vLLM in preference to llama.cpp. darwin-x86 had the identical hole. It was documented as a deliberate omission because nothing accelerates on an Intel Mac, which is true about acceleration and wrong about consequence: with every engine tied, download size decides. Add rows for both putting llama-cpp first. The GPU engines are enumerated behind it rather than left unmatched: an unmatched engine already ranks below every listed one, so llama.cpp would win either way, but unmatched engines also tie with each other and let size decide among them. Naming them fixes that order. MLX is left off the darwin-x86 row on purpose so it ranks last, since IsBackendCompatible admits darwin-tokened engines on that capability even though MLX needs Apple silicon. Preference orders survivors and never filters, so a model published only as a vLLM build is still installed on a host with no GPU; there is a spec for it. Surveyed every other value getSystemCapabilities() can return. nvidia, amd, intel and vulkan have rows; the l4t and cuda-refined values reach the nvidia row by prefix; "apple" and "" cannot reach the vendor fallthrough because the darwin and no-GPU branches return earlier. default and darwin-x86 were the only live holes. BackendPreferenceTokens and its build-tag table are untouched, and preferenceRank stays generic, naming no engine and no capability. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * gallery: prefer speculative-decoding builds when they fit Rank serving features between engine preference and size, so a host that can hold a DFlash or MTP build of a model's weights installs it instead of the plain build. Both answer faster for the same output, so whenever one survives the filters there is no reason to take the plain build. Precedence is now fit, then engine, then serving feature, then size. Engine outranks the feature deliberately: a serving feature makes the right engine faster, it does not make a wrong engine right, so a plain vLLM build still beats a DFlash llama.cpp build on NVIDIA. Fit outranks both, and a drafter pairing is strictly larger than the plain build, so the existing size filter drops it on a host too small for it before this axis is consulted. The order lives in a third preference table in pkg/system, alongside the build tag and engine name tables. It is the odd one of the three: not keyed by capability, because no hardware prefers a plain build over an equivalent faster one, and matched against whole segments of a gallery ENTRY NAME rather than as a substring of a backend value. Nothing on a gallery entry declares a serving feature, and tags are not a usable substitute: gemma-4-e2b-it:sglang-mtp carries an mtp tag while ornith-1.0-9b-mtp and qwen3.6-27b-nvfp4-mtp carry none. Entry names are author-supplied free text, unlike the closed engine vocabulary, so a short marker can turn up inside an unrelated word and whole segment matching is what keeps smtp-assistant from ranking as an MTP build. The block comment over the tables now documents all three together and states what each is matched against; the ranking code names no feature, so adding one stays a one-line edit to the table. 29c49203b rejected these entries as serving configurations rather than alternative builds of the same weights. The definition is now "alternative ways to serve the same model", which includes them, so regroup 14 entries under 12 parents. Judged by the files each entry points at: the qwen3.6, qwen3.5, qwen3 and deepseek pairings are the base GGUF plus a drafter, the gemma-4 QAT MTP entries are the same QAT weights at a different quantization plus an MTP drafter, and the two sglang MTP entries describe themselves as the same model served with speculative decoding. Left separate: qwen3.6-27b-mtp-pi-tune, a finetune with its own weights, and every entry whose base model LocalAI does not ship as its own row, which is the whole Qwopus line plus gemmable-4-12b-mtp, mimo-7b-mtp:sglang and qwen3.5-4b-dflash. None of the twelve parents defines a YAML anchor, so no variants key can leak through a merge key and no empty override was needed this time. The index was edited by line insertion only. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * test: check env restore errors in capability and variant specs errcheck flagged ten unchecked os.Setenv and os.Unsetenv returns in the specs added while the pre-commit hook was being skipped. Restoring an env var is exactly the place a silent failure leaks state into the next spec, so assert on it rather than suppressing the linter. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): make the mtp tag authoritative for serving-feature ranking Variant auto-selection ranks survivors by fit, then engine, then serving feature, then size. The serving-feature lookup read only whole alphanumeric segments of a variant's entry name, because tags were inconsistent: every dflash entry carried a dflash tag, but only 7 of 20 MTP entries carried an mtp tag. Tag the 13 untagged MTP entries, then teach the lookup to read tags as well as names. A tag is now the authoritative signal and is compared whole and case-insensitively, which is safe precisely because a tag is a deliberate declaration rather than free text: there is no word-inside-a-word failure mode, so the segment splitting the name half needs is unnecessary there. The name check stays as a fallback rather than being replaced. Switching to tags only would have regressed the six already-grouped entries on the day it shipped, and would depend on tagging discipline that does not exist yet. The lookup still names no feature, so adding one remains a one-line edit to servingFeaturePreferenceTokens. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): make a declared tag the sole serving-feature signal Variant auto-selection ranks survivors by fit, then engine, then serving feature, then size. The serving-feature lookup recognised a speculative build by either a declared tag or a whole segment of its entry name. Drop the name half: a tag is now the only signal. A name is author-supplied free text and a naming convention is not a contract, so reading a marker out of one infers a capability nobody declared. The gallery already had the failure in it: the four NVFP4 entries name MTP-bearing weights while setting no option that enables speculative decoding, and being live variants they were winning the feature axis without answering any faster. overrides.options was considered as the replacement and rejected. It carries spec_type:draft-mtp / spec_type:draft-dflash, which is what actually turns the feature on, but that spelling is llama.cpp's config vocabulary: ds4 spells the same feature mtp_path and sglang spells it speculative_algorithm in a referenced config. Keying a cross-backend ranking decision on one backend's option syntax would rank the other backends' builds as plain. Options are the curation-time check instead, and never reach the selection logic. With no fallback left, tag correctness is load bearing, so audit every entry against the rule "tagged when the entry configures that feature, in whatever vocabulary its backend uses". Three entries configure MTP untagged and gain the tag (hy3, glm-5.2, qwythos-9b-claude-mythos-5-1m, all spec_type:draft-mtp with no marker in their names). Four carry the tag while configuring nothing and lose it: qwen3.6-27b-nvfp4-mtp, qwen3.6-35b-a3b-nvfp4-mtp, qwopus3.6-27b-coder-mtp-nvfp4 and qwopus3.6-27b-v2-mtp-nvfp4, whose only option is use_jinja:true. The dflash side was checked independently rather than assumed consistent: all five dflash entries declare spec_type:draft-dflash and all five are tagged, so it needed no edits. Four entries keep a tag that a literal spec_type-only reading would strip, because they configure MTP through a different backend: deepseek-v4-flash-q2-mtp via ds4's mtp_path/mtp_draft, and the three sglang entries via speculative_algorithm in their referenced configs. Stripping those would contradict the reason spec_type was rejected as the signal and would demote four genuinely faster builds to plain. The index was edited by line insertion and deletion only, never round-tripped through a serializer. A resolved-tag diff across all 1272 named entries, taken after merge keys are applied, shows exactly these 7 changing and no entry gaining or losing a tag through an anchor. The two specs that pinned the name fallback are inverted rather than deleted, since a name silently promoting a build is the regression worth guarding. The whole-token guard survives on the tag path, where smtp must still not match mtp. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): make deepseek-v4-flash variant targets installable Clicking install on deepseek-v4-flash failed with "invalid gallery model". The parent entry is fine, but all four entries it was grouped with declared neither url: nor config_file:, and applyModel needs one of the two to have anything to build a config from. They carry urls: (plural), the informational HuggingFace link list, which is a different field. None of the four was ever independently installable, so grouping them routed a previously-working install into a broken entry. Give each the url: the parent already resolves through. virtual.yaml is a no-op base, and applyModel passes overrides to InstallModel separately from the fetched config, so backend: ds4, the parameters and the ssd/mtp options all still land exactly as authored. This is the same pattern the parent and many other GGUF entries in the index already use. Add the lint rule that should have caught this. checkVariantReferences only proved a target exists and is not itself a parent, which is structural validity: an entry can exist, declare no variants, and still be uninstallable. checkVariantTargetsInstallable mirrors applyModel's precondition instead, and names the parent, the target and the missing fields, because whoever hits it is reading a gallery entry and has no reason to know applyModel exists. The two index-driven resolution specs live in their own Ordered container: an Ordered container stops at its first failure, so sharing one with the lint rules let a lint breach skip them silently. Nine further entries gallery-wide have the same defect and are unrelated to variants, so they are broken installs that predate this branch. They are left alone here rather than buried in a regression fix, and widening the rule to cover every entry is deferred with them so the gate can ratchet up in one step instead of needing a skip list. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(gallery): install entries with no url or config_file on an empty base applyModel had three branches: fetch a base config from url:, build one from an inline config_file:, or fail with "invalid gallery model". An entry declaring neither is now installed on an empty base config, with overrides: and files: supplying everything. This is what the ~345 entries pointing at gallery/virtual.yaml were already getting. That stub is five lines carrying name, description and license. description and license are overwritten from the gallery entry immediately after the fetch, and the name never reaches disk because InstallModel prefers the install name. Crucially applyModel passes model.Overrides to InstallModel as a separate argument rather than merging it into the fetched config, so nothing an author writes depends on that base existing. The fetch bought a round trip to GitHub and nothing else. That makes f4ef80173 the wrong fix, so it is unwound. The four url: lines it added to the deepseek-v4-flash variants are reverted: they are a pointless network fetch now, and the family installs without them. Relaxing the branch would hide a real authoring mistake, so a payload rule replaces the base-config rule. An entry with no url, no config_file, no overrides and no files installs nothing and would leave an empty model directory while reporting success, so it is refused by name. The caller's request counts toward the payload, because its overrides and files are merged into the install exactly as the entry's own are. urls: (plural) is the informational link list and does not count, which is what the four entries that shipped broken had and why they were still uninstallable. checkVariantTargetsInstallable asserted every variant target declares a url: or a config_file:, which is no longer true and would now reject correct authoring. checkEntriesInstallSomething pins what survives instead, and covers every entry rather than only variant targets: the hazard is a half-written stanza and a parent can be one as easily as a target. The old rule was scoped to targets precisely because nine unrelated entries would have failed a gallery-wide version; those nine are valid now, so the deferred ratchet happens here in one step. 1280 entries, zero violations. Those nine (aurore-reveil_koto-small-7b-it, lfm2-1.2b, the six liquidai_lfm2 entries and deepseek-v4-pro-q2-ssd) become installable for free. Each carries overrides: and files:, and one of them is driven through the real install path in a spec. The no-fetch spec is paired rather than bare: an assertion that nothing was fetched proves nothing unless something could have been, so a control runs the same fixture with a url: pointing at a base config that is not there and asserts the install fails. Only then does the identical fixture without the url passing mean the read was skipped. Follow-up, deliberately not here: the ~345 entries still naming virtual.yaml can drop their url:. That is 345 index edits with their own risk, and mixing them in would bury this change. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ui(models): let search bypass the variant collapse, drop the toggle The models page collapsed the gallery to one row per model by default and offered a toggle to see every individual build. Because the collapse composed with the search term, a build another entry offers as a variant could not be found by typing its name, so the toggle was the only way to reach those builds in the UI. A user who typed a name they knew existed got "no models found", which reads as "that model does not exist". Collapse is for browsing; search is for finding. An explicit search term now bypasses the collapse in the listing handler, so a name lookup returns matching entries whether or not a parent offers them. The term is trimmed once at the top of the handler, so whitespace is neither a search nor a bypass; previously an untrimmed blank term also narrowed the listing to whatever contained a space. Tag and backend deliberately do not bypass: they refine a listing the user is still reading rather than name an entry already known to exist. That makes the toggle redundant, so it goes, along with its i18n strings in all six locales, its localStorage persistence, its participation in "Clear filters" and the empty-state hint telling users to turn it off. The hint was doubly stale: it pointed at a control that no longer exists, and it was untrue exactly when a user has a search term, since searching now sees every build. The page always requests the collapsed listing. The stored preference key is left inert rather than cleaned up: nothing reads it, so a user who had the toggle off simply gets the collapsed view. collapse_variants stays on the API, off by default, because other clients want either view and the UI dropping its control is no reason to remove a working parameter. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): give the models gallery filter form a deliberate structure The filter area had accreted controls into one undifferentiated flow. The "Fits in GPU" toggle and the backend select were direct children of .filter-bar, the same wrapping container as the 18 taxonomy chips, so their position was decided by how many chips happened to wrap at the current width rather than by any layout intent. At narrow widths they were pushed past the right edge of that container's horizontal scroll and became unreachable entirely. Restructure into three bands inside the house .filter-bar-group wrapper that components/FilterBar.jsx already uses on Backends and the System tabs: 1. query scope: search plus the backend select 2. taxonomy: the chip row, alone, free to wrap 3. refinements: fits-in-GPU and context size, under a hairline rule The backend select leads the chips rather than trailing them because picking a backend disables the use cases that backend cannot serve, so it gates the row below it. Fits-in-GPU and context size share a band because they are one control group: the context size is the length the VRAM estimate is computed at, and that estimate is what the fits filter tests against. Chips had no visible keyboard focus indicator. The global focus ring is wrapped in :where(), so it carries the specificity of a bare :focus-visible, ties with .filter-btn and loses on source order, leaving focused chips showing their resting drop shadow. Restate the ring where it outranks both resting and hover. Also: aria-pressed on the chips, a real label association and aria-valuetext on the context slider (it steps over an index, so it announced "2"), disabled chip styling moved off inline styles, a prefers-reduced-motion block for the chip transition, and the hard-coded English "Context:" moved into all seven locales. No behaviour change: same filters, same state, same requests. Page reset on change, localStorage persistence and "Clear filters" verified unchanged. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): let the models recommendations panel fade into the background The "Recommended for your hardware" strip rendered at full height on every visit regardless of how many models were already installed, costing 186px at 1600px wide (287px at 1100px, where its cards wrapped to two rows) and pushing the first gallery row to y=554 / y=703. Make its prominence track how much the user still needs it. The panel now defaults to a one-line summary once anything is installed, and both the collapse choice and the existing dismissal persist: collapsed = explicit user choice, if one exists : installedCount > 0 The preference is three-valued on purpose. A boolean cannot tell "the user expanded it" apart from "the user has never chosen", and those need opposite handling when the installed count later crosses zero: someone who deliberately opened the panel on an empty instance should not have it collapse out from under them when their first model finishes installing. Collapsed keeps the card, icon, title and a suggestion count, so the panel is recovered by clicking what you are already looking at rather than by hunting. Expanded is unchanged, because for a user with nothing installed it was never the problem. Collapsed reclaims 145px at 1600 and 420, and 246px at 1100. Models.jsx gains a statsLoaded flag: stats initializes to installed:0, so reading it before the fetch resolves would render expanded and collapse a frame later, which is exactly the layout shove this removes. The dismissal key moves to the page's localai-models-* convention; the old localai_rec_models_dismissed is still read, never written, so an existing dismissal is honoured rather than resurrected by the rename. Accessibility: the disclosure is a real button whose accessible name is the visible title alone, with state on aria-expanded and aria-controls resolving in both states, because the grid is hidden via the hidden attribute rather than unmounted. That also keeps the four install buttons out of the tab order while collapsed. The app's global focus ring applies; no per-component outline is added, per the warning in App.css. Reveal animates opacity and transform only, never height, and both it and the chevron rotation are disabled under prefers-reduced-motion. Only en had a recommended block, so the other six locales were falling back to English for the whole panel. Translated the complete block rather than adding one orphaned key to files that would still render the title in English. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(downloader): recover from a leftover .partial on non-HTTP URIs An interrupted download leaves a `<file>.partial` behind. The partial handling in DownloadFileWithContext gated resume on `err == nil && uri.LooksLikeHTTPURL()`, so for any URI that is not literally http(s) the branch fell through to `else if !errors.Is(err, os.ErrNotExist)`, which with a nil err is true. The download then failed with an error wrapping nil: failed to check file ".../Ternary-Bonsai-27B-Q2_g64.gguf" existence: <nil> Every gallery file URI uses `huggingface://`, so a single interrupted download made that model permanently uninstallable until someone deleted the partial by hand. The `<nil>` in the message compounded it by pointing debugging at a filesystem failure that never happened. Restructure the handling as an explicit switch over the four real states: partial exists and is resumable, partial exists and is not resumable (discard and restart, as already done for an HTTP server without range support), no partial, and a genuine stat failure. The error branch is now only reachable with a non-nil error, names the path that was actually stat'd, and wraps with %w. Discarding is required for correctness and not merely convenience: the writer opens the partial with O_APPEND, so an un-resumed download would concatenate a fresh body onto stale bytes. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): tell the models gallery's variant rows apart, and let browsing see every build Both variant surfaces rendered name, backend and size. For two builds of one model that is close to no information: a variant exists precisely because the same weights are offered another way, so the backend usually matches and the sizes usually land within a few hundred megabytes. Comparing ternary-bonsai-27b-pq2 against ternary-bonsai-27b-q2-g64 meant reading two names that differ by a suffix nobody has defined anywhere in the UI. Report the quantization and the serving features on VariantView, and derive both server-side from the referenced entry rather than parsing names in the browser, so every client reads the same format out of the same file the installer will hand the backend. Quantization comes from overrides.parameters.model first, falling back to the file list. That order is load bearing: entries routinely ship a vision tower alongside the language model at a different quantization, so reading the file list first reports the mmproj's format. Matching walks `-` and `.` delimited segments right to left; `_` deliberately does not split, because it separates the parts INSIDE a quant token and splitting on it reports Q4 for a Q4_K_M build. A second, looser pass takes a segment's `_`-delimited tail, which catches the gemma-4-E2B_q4_0-it.gguf style; it runs second so a precise match can never lose to a fuzzy one further right in the name. An entry naming no format reports nothing, which is the honest answer for a backend served from a directory of weights. Features are the same tag-against-vocabulary match servingFeatureRank already ranks on, over the same host preference list. A build can therefore never be shown as faster than one selection did not actually reward, nor rewarded without being shown; a spec pins that agreement rather than trusting it. The compact dropdown gets the quantization on its meta line and the bare feature token. The detail row, which has the room, gets the quantization as its own monospaced column so precision lines up down the list, and the feature spelled out, because DFLASH names nothing to a user who has not met it. The referenced entry's description stays out of both: the detail row already renders the parent's prose above the table, and a second block per variant would push a three-variant list past a screen to restate what the columns now say precisely. The collapse toggle comes back. 462583f38 dropped it once search bypassed the collapse, on the reasoning that nothing was unreachable any more. That holds for finding a build whose name you know and does not hold for browsing: no sequence of actions enumerated the 68 builds the default view hides. Collapse is for browsing and search is for finding, and the toggle was the browsing half. It goes in the refinements band 0d4823362 established, not back among the taxonomy chips where its position depended on how many chips happened to wrap. It leads that band because it decides how many rows the other two refine over, and because unlike fits-in-GPU it is unconditional: a host with no GPU still browses. The search bypass is untouched and re-checked by a spec in the toggle's default state, since restoring the control must not restore the dead end it replaced. The empty-state hint returns but only without a search term, because a term bypasses the collapse and the hint would otherwise point at a control that cannot change the result. The stored preference reads 'on'/'off' only: an older build wrote '1'/'0' from an effect that ran on mount, so those record that the page was opened, not that anyone chose a view. Also fixes a latent flake it exposed. The collapse_variants spec compared whole response bodies byte for byte, and the listing envelope carries live host telemetry that drifts between two calls milliseconds apart, so it was asserting on the machine's memory pressure. It now compares everything the parameter governs -- the entries, their serialization and the paging -- and is green 25/25 where it was failing about one run in three. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): let the models gallery show a variant's full details The variant list in an entry's expanded detail row says how the builds differ: name, backend, quantization, size, and the auto-selected, base and serving-feature markers. It cannot say what any one of them is. A variant's own description, tags, license, source links and file list are unreachable anywhere in the UI, because while the collapse is on a variant has no gallery row of its own. Give each variant row an info control that reveals its entry, rendered by the same ModelDetail a top-level row gets, so a field added to the detail view appears here too. variantData is withheld from the nested render: a variant may declare variants of its own, and recursing would nest a picker inside a picker two levels deep already. An inline disclosure rather than a modal. The control sits inside a table row that is already expanded, inside a variant list within that; a dialog opened from there stacks a dismissal on a dismissal for a handful of extra fields about the entry the user is already reading, and breaks the page's own expand idiom. The third level is carried by an inset and a left rule instead of another card. The entry is fetched by exact name from the listing, once, on first use. The listing already returns every field the detail view renders, and a search term bypasses the variant collapse server-side, so no new endpoint is needed and neither the listing nor DescribeVariants gains any work. Expanding a row costs nothing; a variant nobody opens costs nothing. A name the listing no longer returns is stated, not blanked: an empty panel reads as a rendering fault rather than as a lookup that came back empty. The control is a sibling of the install button, not a descendant, so asking about a build can never install it. The variant list keeps its content-sized columns via a trailing filler track instead of max-content sizing, so the rows are unchanged while the panel spanning them gets the pane width its file table needs. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(gallery): let search respect the collapse instead of switching it off The models listing collapsed to one row per model, and an explicit search term turned that off wholesale. Searching while collapsed therefore answered with the individual builds a parent already offers, which are exactly the rows the view the user asked for has no place for: typing "mtp" returned qwen3.6-27b-nvfp4-mtp, a row that is invisible the moment the box is cleared. The bypass was the right shape of fix for the wrong half of the problem. What a search must not do is answer "no models found" for a build the gallery does hold; that does not require abandoning the grouping the user asked for. So the term is now matched against every entry either way, hidden builds included, and the collapse decides how a match is reported rather than which matches exist. Collapsing stops being a filter that drops rows and becomes a substitution: a match on a build another entry offers is reported as that entry, the one installable in its own right. Nothing becomes unfindable and nothing comes back that the requested view cannot show. Substitution happens after search, tag and backend, so every filter is judged against the build that really carries the name, tag or backend rather than against a parent that merely offers it; the other order would let backend=vllm match a parent whose own backend is something else. The price is that the surfaced row shows the parent's own metadata while the match was on a variant, which is what grouping means, and the alternative is claiming the gallery holds no such build. It happens before the count and the page math, so both describe the rows actually handed out rather than the matches that produced them. A parent already in the result keeps its own position and absorbs its matching variants there, which is what leaves the browsing listing ordered exactly as it was; a parent surfaced only by a variant takes the position of the first variant that surfaced it. Either way it appears once, however many of its builds matched and whether or not it matched itself. Search preserves gallery order rather than scoring, so a surfaced parent has a real position rather than an invented one. VariantParents never reports an entry that declares variants of its own, so a parent is never itself hidden and one hop always lands on a visible row. The handler follows exactly one anyway: refusing the second is what makes a gallery the linter would have rejected terminate rather than loop. The empty-state hint pointing at the toggle goes with it for every server-side filter. Substitution means a match is always reported as some row, so the collapse can no longer be why a term, a chip or a backend came back empty, and naming it there sends the user to a control that cannot change the result. It survives for the fits filter alone, which runs in the browser after the substitution and judges the surfaced entry's own size: there the build that fits really can be filtered out along with a parent that does not. Searching a build's exact name while collapsed now answers with its parent, so the result no longer contains the string the user typed. That is intended, and the row is the one they can act on, but it is a real rough edge: nothing on the row explains the connection. Closing it properly means reporting which variant matched so the UI can say so, which the listing does not do today. ResetGalleryModelCache is added for tests. The model cache is a package global keyed by nothing, so a background refresh one spec triggers can land in the middle of the next and answer it with the previous spec's gallery; the extra specs here made that fail about one run in five. It waits for the in-flight refresh to publish before clearing, since clearing alone only narrows the window. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-07-20 18:43:02 +02:00
LocalAI [bot]	b00422e45f	feat(backends): add LongCat video and avatar generation (#10792) ... * feat(backends): add LongCat video and avatar generation Assisted-by: Codex:GPT-5 [apply_patch] [exec_command] [web] * refactor(config): declare model I/O modalities Make model configs declare input and output modalities so capability discovery no longer branches on backend or checkpoint names. Complete the LongCat gallery and user documentation, make the SDPA patch apply to the pinned upstream revision, and stabilize the Agent Jobs race exposed by the required hook. Assisted-by: Codex:GPT-5 [web] --------- Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-07-12 23:58:46 +02:00
LocalAI [bot]	40dae953f4	feat: interleaved thinking with tool calls (reasoning_content alias + Anthropic thinking blocks) (#10744) ... * feat(schema): accept reasoning_content as inbound alias for reasoning Interleaved-thinking clients (cogito, vLLM/DeepSeek-style) emit reasoning_content on assistant turns. Accept it as an inbound alias so reasoning survives the tool-result loop; canonical reasoning wins when both are present. Emission is unchanged (still reasoning). Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * test(schema): pin interleaved reasoning+tool_calls round-trip Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * test(openai): pin reachedTokenBudget truncation detection Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(anthropic): add thinking and signature fields to content blocks Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(anthropic): parse inbound thinking blocks into reasoning Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(anthropic): emit thinking blocks with synthetic signature on tool turns Extract buildAnthropicContentBlocks so non-streaming content assembly is unit-testable, and prepend a thinking block (with an opaque synthetic signature) before text/tool_use blocks when the request opts into thinking. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(anthropic): stream thinking_delta and signature_delta before tool_use Extract anthropicStreamSequence so the streaming block order is unit-testable, and emit content_block_start(thinking) -> thinking_delta -> signature_delta -> content_block_stop before the tool_use block sequence when thinking is enabled. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * docs: add interleaved thinking with tool calls guide Add a features guide describing interleaved thinking: an assistant turn carrying reasoning and tool_calls together, the reasoning-round-trip contract (including the reasoning_content inbound alias and Anthropic thinking blocks with a synthetic signature), per-backend enablement (reasoning_format for llama.cpp, reasoning_parser/tool_call_parser for vLLM/SGLang plus the vLLM auto-config hook), a worked request/response example, and known limitations. Cross-link from model-configuration, text-generation, and openai-functions. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-07-08 16:45:43 +00:00
LocalAI [bot]	b0959d4756	feat(api): add GET /v1/models/capabilities endpoint (#10687) ... Additive superset of /v1/models that enriches each model entry with the capabilities it supports plus its input/output modalities (text / image / audio / video). Clients that only understand /v1/models are unaffected -- they simply never call the new route. Audio and video *input* are derived from the model's multimodal limits (vLLM limit_mm_per_prompt), which no single usecase FLAG expresses. That gap is exactly why a plain capability list is insufficient and this enriched endpoint exists: an attachment router can now decide whether an image/audio/video file can go to the active model directly, or must be converted/transcribed first. Capability derivation lives in core/config as the single source of truth (ModelConfig.Capabilities / InputModalities / OutputModalities / VisionSupported / ...); the Ollama capability surface now delegates to it instead of keeping a parallel copy. Vision is gated on chat/completion capability so a MediaMarker hydrated onto a non-chat model (e.g. a pure ASR/TTS backend) no longer reports a false vision capability. Read-only listing: no new FLAG_* flag, reuses the existing `models` swagger tag, and intentionally exposes no MCP admin tool (there is nothing to manage conversationally). Assisted-by: Claude:claude-opus-4-8 [Claude Code] Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-07-05 08:51:55 +02:00
Richard Palethorpe	eb32cd9073	feat(realtime): eager blocking pipeline warm-up + /backend/load API (#10662) ... Realtime sessions previously lazy-loaded each pipeline sub-model (VAD, transcription, LLM, TTS) on first use, so every cold session paid a per-request model-load stall and load errors only surfaced mid-stream. Warm the whole pipeline eagerly and blockingly at session start (including the voice-gate speaker-recognition model, which an enforced gate blocks each utterance on; compaction's summary_model stays lazy since it only runs off the response path): - Add backend.PreloadModel / PreloadModelByName as the single load path for every modality (no transcription special-case; backend-omitted configs are deprecated). - The realtime session blocks on Model.Warmup and returns a model_load_error to the client if any stage fails to load; updateSession warms in the background. Opt out per pipeline with pipeline.disable_warmup, exposed as a UI toggle via the config-metadata registry. Add a LocalAI-native POST /backend/load (and /v1/backend/load) that pre-loads a model -- expanding realtime pipelines into their sub-models -- as the inverse of /backend/shutdown. There is one preload engine (backend.PreloadStages): the realtime Warmup methods, /backend/load and the --load-to-memory startup flag all use it, so --load-to-memory now also expands pipeline models and records load-failure traces. Pipeline sub-model alias resolution is likewise shared (ModelConfigLoader.LoadResolvedModelConfig). Surface the endpoint everywhere an admin manages models: - MCP admin tool load_model (httpapi + inproc clients, safety/catalog prompts, catalog/dispatch tests). - "Load into memory" action in the React models UI. - Swagger regenerated; docs moved to the general backend-monitor page since it is not realtime-specific. Fix a Traces UI crash ("json: unsupported value: -Inf"): audio-snippet RMS/peak now floor at a finite dBFS, and backend-trace data is sanitized to drop non-finite floats before marshaling. The sanitizer is copy-on-write -- it runs on every RecordBackendTrace, so containers are only re-allocated on the paths that actually changed. Migrate core/http/openresponses_test.go onto the prebuilt mock-backend the rest of the http suite already uses -- it was the last spec still pointing at a real HuggingFace model, so it 404'd wherever no vision backend was built -- and fix its item_reference specs to send the spec's "id" field instead of "item_id", which the handler never accepted. Assisted-by: Claude:claude-opus-4-8 Claude Code Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-07-03 18:00:37 +02:00
Richard Palethorpe	5d0c43ec6e	feat(realtime): Semantic VAD EOU token (#10444) ... * feat(realtime): EOU-driven semantic_vad turn detection Add a `semantic_vad` turn-detection mode to the realtime API that feeds the transcription model live and decides "the user finished speaking" from the `<EOU>` end-of-utterance token rather than from silence alone. When EOU fires the turn commits immediately (~0.3s); otherwise it falls back to an eagerness-scaled silence threshold (low/med/high = 8/4/2s). Plumbing, bottom to top: - proto: `AudioTranscriptionLive` bidirectional RPC (config-first oneof, mono float PCM @16k, ready-ack / Unimplemented degrade signal) plus `TranscriptResult.eou` for the unary retranscribe gate. - pkg/grpc: client/server/base/embed scaffolding for the bidi stream, modeled on AudioTransformStream; release stream conns on terminal Recv. - parakeet-cpp: live transcription RPC with per-C-call engine locking (one live stream per turn, finalize+free at commit); bump parakeet.cpp to ABI v5 — incremental StreamingMel (no more quadratic per-feed mel recompute that delayed EOU on long turns) and the <EOU>/<EOB> split; strip the literal <EOU>/<EOB> from offline text and set Eou. - core/backend: LiveTranscriptionSession wrapper + pipeline `turn_detection:` config block (type/eagerness/retranscribe). - realtime: semantic_vad integration — live input captions streamed as transcription deltas while the user speaks, EOU-immediate commit with eagerness fallback, optional retranscribe gate (batch re-decode must also end in <EOU> to confirm), clause synthesis off the LLM token callback, and per-turn live-transcription / model_load telemetry. - UI: show the realtime pipeline components as a vertical list. Docs and tests included; opt-in via the pipeline YAML or per-session `session.update`. Non-streaming STT backends degrade to silence-only. Assisted-by: Claude Code:claude-opus-4-8 [Read] [Edit] [Write] [Bash] Assisted-by: Claude Code:claude-fable-5 [Read] [Edit] [Bash] Signed-off-by: Richard Palethorpe <io@richiejp.com> * feat(realtime): explicit formally-verified state machines + parakeet streaming driver The realtime API had several implicit state machines whose state was inferred from scattered booleans, channels, and five separate mutexes, leaving illegal/inconsistent states reachable. Make them explicit and keep the implementation in step with a formal design; rework the parakeet streaming backend along the same lines. Realtime state machines (M1-M5). Each is a sealed sum-type State/Event/Effect with a total, pure Next(state,event)->(state,[]effect) behind a single-writer Coordinator: M1 conncoord connection lifecycle: VAD toggle + once-only teardown (replaces vadServerStarted + a `done` channel closed from two sites). M2 turncoord turn detection: collapses speechStarted and the live-stream "turn open" flag into one state, so discardTurn can no longer desync them and suppress the next onset. M3 respcoord response coordination: serializes the dual-writer start/cancel so at most one response is live; one response.done per response.create. M4 compactcoord conversation compaction: single-flight (replaces the `compacting atomic.Bool` CAS). M5 ttscoord TTS pipeline: open->closing->closed, idempotent wait(), rejects enqueue-after-close (was a silent drop). The Coordinator/Sink/Next plumbing — only the sealed types and Next differed per machine — is extracted once into core/http/endpoints/openai/coordinator as a generic Coordinator[S,E,F]; each machine keeps its public API via type aliases, so no sink, call-site, or test moved. Hierarchy. session_lifecycle.fizz models M1 as the parent region with its children (M2/M3/M4) as one statechart and asserts ChildrenDieWithParent (conn torn => all children terminal, none start after teardown). respcoord and compactcoord gain an absorbing Terminated state + Shutdown event; conncoord's teardown drives the children terminal. This closes a compaction teardown gap: a fire-and-forget compaction could outlive a torn session — compactionSink now takes a session-scoped cancellable context + WaitGroup and joins the in-flight summarize+evict on shutdown. Formal verification. formal-verification/ holds one authoritative FizzBee spec per machine plus the composition spec, each with an always-assertion and a documented one-line edit that makes the checker fail (verified non-vacuous). scripts/realtime-conformance.sh is fail-closed: all Go conformance suites under -race AND a model-check of every .fizz spec; a missing FizzBee is a hard error (only the loud REALTIME_CONFORMANCE_SKIP_FIZZBEE=1 bypasses it, never in CI). FizzBee is pinned by sha256 and installed via scripts/install-fizzbee.sh into .tools/ (gitignored). Wired as make test-realtime-conformance, a CI workflow, and a pre-commit path filter. Go conformance tests are Ginkgo/Gomega (per the repo's forbidigo lint): transition tables + fixed-seed property walks + concurrent/-race specs, no rapid dependency. Design map: docs/design/realtime-state-machines.md. Parakeet streaming backend. The same treatment applied to the parakeet-cpp streaming paths: - AudioTranscriptionStream returns codes.Unimplemented for non-streaming models instead of decoding offline and emitting it as one delta + final. A client that asked for streaming learns the model cannot stream rather than receiving a batch result shaped like a stream. New grpcerrors.StreamTranscriptionUnsupported carries that signal; the HTTP /v1/audio/transcriptions stream path surfaces it as an SSE error event. Mirrors AudioTranscriptionLive, which already did this. - utteranceBoundary (boundary.go): a single definition of the end-of-utterance latch, replacing three open-coded finalEou toggles. Modelled as a two-valued type so illegal states are unrepresentable. - Shared decode driver (driver.go): streamFeedResult (one per-feed event) + feedChunk (hides the ABI v4 JSON vs text-only split) + feedSlices + flushTail. The feed loop is written once. - AudioTranscriptionLive becomes a bidi adapter: it streams the per-feed {delta,eou,eob,words} the realtime turn detector consumes and a terminal FinalResult carrying only Text. Segments/duration/eou are offline-only and no longer produced (nor read) on the live path; liveTraceState drops the terminal eou and keeps the per-feed eou_events count. - AudioTranscriptionStream + streamJSON merge into one driver-based function; streamSegmenter is generalized to the unified event with a text-only fallback that preserves the legacy (no-words) library's per-utterance segmentation. Verified: build/vet/gofumpt clean, golangci-lint 0 issues, all coordinator and parakeet packages under -race, the fail-closed conformance gate green, and make test-realtime (12 e2e WS+WebRTC). Assisted-by: Claude:claude-opus-4-8 [Claude Code] Signed-off-by: Richard Palethorpe <io@richiejp.com> --------- Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-06-30 09:01:22 +02:00
LocalAI [bot]	f0d0bff232	fix(llama-cpp): stop reinterpreting plain-string message content as JSON (#10524) (#10538) ... The llama-cpp gRPC backend reconstructs OpenAI messages from proto for the tokenizer-template path and blindly json::parse'd each message's content string. LocalAI's Go layer always flattens content to a plain string, so a user prompt that merely looks like JSON (e.g. mealie's ingredient array ["1/4 cup brown sugar", ...]) was reinterpreted as structured content parts and rejected by oaicompat_chat_params_parse with "unsupported content[].type". Normalize content per role instead: user/system/developer content is opaque text and is never JSON-sniffed; assistant/tool content still collapses a literal JSON null/object (tool-call bookkeeping) to a string, but a plain string is never turned into an array/scalar. The array defense is role-independent, so the role gate only governs the benign null/object case. While here, extract the duplicated per-message reconstruction and the pre-template content sanitization into shared, unit-tested helpers (message_content.h) so the streaming (PredictStream) and non-streaming (Predict) paths cannot drift. This removes ~490 lines of copy-pasted defensive code, the dead tool-role parse branches, and the redundant Predict-only tool_calls branch, while preserving the prior #7324 (null content -> "") and #7528 (tool array content -> string) fixes. Tests: - backend/cpp/llama-cpp/message_content_test.cpp: standalone C++ unit tests for all three helpers (#10524, #7324, #7528, multimodal), discovered and run by `make test-backend-cpp` and a new generic tests-backend-cpp CI job. Also wired as an opt-in CMake/ctest target (-DLLAMA_GRPC_BUILD_TESTS=ON). - core/schema/message_test.go: Go regression pinning that ToProto flattens a JSON-array-looking text part to the verbatim string. - prepare.sh now copies message_content.h into the build tree. Assisted-by: Claude:claude-opus-4-8 [Claude Code] Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-06-27 01:42:05 +02:00
LocalAI [bot]	600dafd20b	feat(ced): sound-event classification backend (CED audio tagger) (#10425) ... * feat(ced): sketch sound-classification backend (CED audio tagger) Wires ced.cpp (CED, 527-class AudioSet sound-event tagger; baby cry, footsteps, glass, alarms, dog bark) into LocalAI as a Go/purego backend. SKETCH (backend skeleton real; core REST wiring + CI/gallery is a checklist in DESIGN.md): - backend/backend.proto: new SoundDetection rpc + SoundClass messages (run `make protogen-go` to regenerate pkg/grpc/proto). - backend/go/ced: main.go (purego dlopen libced.so + ced_capi.h), goced.go (Ced gRPC backend: Load + SoundDetection), Makefile (clone-at-pin CED_VERSION, ggml static-PIC shared build), run.sh, package.sh, .gitignore. - DESIGN.md: REST /v1/audio/classification wiring (handler/route/capability registration checklist), gallery/index + CI registration, and a scoping note for the realtime/websocket live-recognition path (sliding-window classify over the existing ws transport + voicegate; the ced C-API per-PCM entry point is already window-friendly). Backend code does not compile until protogen-go regenerates the pb types and a libced.so is built (Makefile clones+builds it). Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): REST /v1/audio/classification endpoint + capability registration Wires the ced sound-event classification backend (AudioSet audio tagger) end to end through the REST surface, mirroring the transcription path. - Handler: core/http/endpoints/openai/sound_classification.go parses the multipart audio upload, temp-files it, resolves the model config and calls the SoundDetection RPC; returns {model, detections[]} JSON. - Backend wrapper: core/backend/sound_classification.go (ModelSoundDetection) loads the model and normalizes the proto response into schema types. - Schema: core/schema/sound_classification.go (SoundClassificationResult). - gRPC layer: SoundDetection wired through the LocalAI wrapper (interface, Backend client, Client, embed, server, base default) so the loader-typed client exposes the RPC; proto regenerated via make protogen-go. - Route: POST /v1/audio/classification (+ /audio/classification alias) with the audio/multipart default-model middleware in routes/openai.go. - Capability surfaces: swagger @Tags/@Router on the handler; FLAG_SOUND_ CLASSIFICATION usecase flag + UsecaseSoundClassification + UsecaseInfoMap + GuessUsecases + ModalityGroups + GetAllModelConfigUsecases; meta usecase option; /api/instructions audio area updated; auth RouteFeatureRegistry + FeatureAudioClassification (APIFeatures, default ON) + FeatureMetas; UI usecaseFilters, capabilities.js CAP_SOUND_CLASSIFICATION, Models.jsx filter + i18n; docs page features/audio-classification.md + whats-new + crosslink. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): realtime sound-event detection over the websocket API When a realtime pipeline configures a sound-classification model, each VAD-committed utterance (the same window the transcription path produces) is also run through the CED sound-event classifier and the scored AudioSet tags are emitted as a new server event. No new backend rpc is needed: the SoundDetection gRPC method already exists on this branch. - config: add Pipeline.SoundDetection (yaml/json sound_detection,omitempty) beside Transcription/VAD. - realtime: add Model.SoundDetection(ctx, audio, topK, threshold) to the ModelInterface; implement it on wrappedModel and transcriptOnlyModel by calling backend.ModelSoundDetection with the session's sound-classification model config (mirrors how Transcribe dispatches). Load the optional config in newModel / newTranscriptionOnlyModel; nil config keeps it additive. - types: add ConversationItemSoundDetectionEvent (item_id, content_index, detections[]{label,score,index}) with type conversation.item.sound_detection, its ServerEventType constant and MarshalJSON, mirroring the transcription completed event. - realtime: add emitSoundDetection (unary path: classify the committed window, build the event, t.SendEvent) and wire it at the utterance-commit hook right after emitTranscription; gated on session.SoundDetectionEnabled (resolved from Pipeline.SoundDetection at session setup, defaults top_k=5, threshold=0). Its error is logged via xlog but never aborts the turn. - test: Ginkgo specs for emitSoundDetection (tags emitted, empty detections, classifier error) plus a SoundDetection method on the fakeModel double. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(ced): implement SoundDetection in nodes backend test doubles The SoundDetection method added to the grpc backend interface left two test doubles (fakeBackendClient, fakeGRPCBackend) incomplete, so core/services/nodes failed to compile under `go vet`/`go test` (go build missed it: the doubles live in _test.go). Add the method to both, mirroring their existing Detect mock. Repairs CI for the nodes package. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): decouple realtime sound detection from VAD (sound-only sessions) Sound-event detection must activate on sounds, not speech, so it no longer runs through the voice VAD/transcription path. A sound-detection-only pipeline (sound_detection set, no transcription/LLM) now: - is accepted by prepareRealtimeConfig (sound_detection counts as a pipeline stage), - builds a lightweight model via newSoundDetectionOnlyModel (no VAD/STT/LLM/TTS loaded), and - defaults the session to turn_detection none (no VAD) with no transcription stage, so the client drives windowing via input_audio_buffer.commit (option A: client-side sliding window). The per-PCM C-API already supports arbitrary windows. commitUtterance gains a sound-only branch: it emits the conversation.item.sound_detection event (scored AudioSet tags) and stops - no transcription, no LLM response. generateResponse is now guarded on a transcription stage being present, so a sound-only turn never invokes the LLM. Existing transcription/VAD sessions are unchanged (additive). Added a commitUtterance sound-only Ginkgo spec asserting it emits the sound event and neither transcribes nor generates a response. go vet + golangci-lint (new-from-merge-base) clean; openai suite green. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): register sound-classification backend in gallery + CI Mechanical backend-image registration for the ced sound-event classifier, mirroring the parakeet-cpp Go/purego backend everywhere it is wired up. - .github/backend-matrix.yml: add the ced build matrix, field-for-field copies of the parakeet-cpp entries (cpu amd64/arm64, cublas cuda 12/13 amd64, l4t cuda-13 arm64, l4t-jetpack cuda-12 arm64, sycl f32/f16, vulkan amd64/arm64, rocm hipblas, and the metal darwin entry), changing only backend and tag-suffix. dockerfile stays ./backend/Dockerfile.golang. - backend/index.yaml: add the &ced meta anchor (capabilities map per platform) plus ced-development and the per-arch image entries, each uri/mirror tag-suffix matching the matrix exactly. The model gallery (GGUF) entry is intentionally deferred pending the HuggingFace publish (TODO note inline). - scripts/changed-backends.js: add an explicit item.backend === "ced" branch in inferBackendPath mapping to backend/go/ced/, same mechanism and ordering as the parakeet-cpp branch (before the generic golang fallthrough). - .github/workflows/bump_deps.yaml: register mudler/ced.cpp -> CED_VERSION in backend/go/ced/Makefile so the daily bot bumps the pin. - swagger/{docs.go,swagger.json,swagger.yaml}: regenerated via make swagger so the existing /v1/audio/classification annotations land in the generated spec. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): server-side windowing for realtime sound detection (option B) Adds an optional server-driven sliding-window classifier so a sound-only realtime client only has to stream audio (no input_audio_buffer.commit): - Pipeline.sound_detection_window_ms / sound_detection_hop_ms config knobs. When both > 0 on a sound-only session, the server classifies the last window of streamed audio every hop and emits a conversation.item.sound_ detection event; the input buffer is trimmed to one window so a long stream stays bounded. When unset, the session stays client-driven (option A). Runs independent of VAD (sound events are not speech). - handleSoundWindow (ticker) + classifySoundWindow (one tick, extracted so it is unit-testable) + writeWindowWAV, which declares the true InputSampleRate (NewWAVHeaderWithRate) so the classifier resamples correctly. Goroutine is started after toggleVAD and torn down with the session (close + wg.Wait). - Register pipeline.sound_detection (+window_ms/hop_ms) in the config meta registry; the earlier realtime commit added pipeline.sound_detection without a registry entry, failing TestAllFieldsHaveRegistryEntries. This fixes that and covers the two new knobs. Tests: classifySoundWindow emits an event + trims the buffer to one window, no-ops on too-little audio; writeWindowWAV declares the given sample rate. go build/vet + golangci-lint (new-from-merge-base) clean; config + openai suites green. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): add ced-base GGUF model gallery entries (f16 + q8_0) The ced-base weights are now published at mudler/ced-base-gguf (Apache-2.0, converted from mispeech/ced-base). Adds gallery/ced.yaml (backend: ced + known_usecases: sound_classification) and two gallery/index.yaml entries (ced-base-f16 default, ced-base-q8 smallest) with sha256-pinned files, and removes the now-resolved TODO from backend/index.yaml. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ced): add tiny/mini/small GGUF model gallery entries Publishes the rest of the CED family (same architecture, metadata-driven port verified end-to-end on ced-tiny) to mudler/ced-{tiny,mini,small}-gguf and adds their f16 + q8_0 gallery entries: ced-tiny (5.5M, edge/Pi-class) f16 11MB / q8_0 6MB ced-mini (9.6M) f16 19MB / q8_0 11MB ced-small (22M) f16 42MB / q8_0 23MB All sha256-pinned. ced-base remains the accuracy default. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore(ced): point gallery entries at the consolidated mudler/ced-gguf repo All CED quantizations (tiny/mini/small/base, f16/q8_0) now live in a single HuggingFace repo, mudler/ced-gguf, instead of per-model repos. Repoint the 8 gallery model entries' urls + file uris accordingly. sha256 and filenames are unchanged. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore(ced): bump CED_VERSION to the short-clip fix Pin the ced backend to ced.cpp 99c6ed3, which fixes a crash on any clip shorter than target_length (~10.11s): time_pos_embed was added at its full 63-frame grid instead of being sliced to the clip's actual time grid, tripping ggml_can_repeat in ggml_add. Surfaced by the live realtime e2e (sub-10s windows) and gated with a short-clip parity test upstream. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * docs(ced): list ced.cpp as a LocalAI-team engine + backend-guide directive - README.md: add ced.cpp to the "native C/C++/GGML engines developed and maintained by the LocalAI project" table. - docs/content/features/backends.md: add a Sound Classification backend category (sound-event classification / audio tagging) listing ced.cpp. - .agents/adding-backends.md: add a "Documenting the backend" section and two verification-checklist items requiring new backends to be documented in the backends.md category list, and in-house native engines to be added to the README maintained-engines table. This directive was missing. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore(ced): repin CED_VERSION to the v0.1.0 release commit ced.cpp history was squashed into a single release commit (tagged v0.1.0), so the previous pin (99c6ed3) no longer exists upstream. Pin to c04ac14, the v0.1.0 release commit, so the backend builds against a commit that exists. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(ced): silence gosec G304/G103 + govet unsafeptr on audited paths - sound_classification.go: os.Create(dst) where dst = temp dir + path.Base of the upload (no traversal). #nosec G304, matching the depth-anything-cpp handler. - goced.go: reading a NUL-terminated C string from a libced-owned buffer. #nosec G103 (gosec) + //nolint:govet (golangci-lint's unsafeptr check), since the uintptr is a C-owned malloc'd buffer, not Go-GC memory. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-06-22 01:00:28 +02:00
Richard Palethorpe	3fa7b2955c	feat(pii): NER tier engine — privacy-filter.cpp backend + NER-centric PII filter (#10360) ... Squashed feat/pii-ner-tier-engine rebased onto master (was 45 commits; see backup/pii-ner-tier-engine-prerebase). Net change: - privacy-filter.cpp: standalone GGML engine for the openai-privacy-filter PII/NER token classifier, wired as a LocalAI gRPC backend (CPU/CUDA/Vulkan). TokenClassify moves off the patched llama.cpp path onto this backend. - PII filter reworked to be NER-centric (encoder/NER detection tier scanning whole conversations as one document), with a recreated bounded restricted- regex secret-matching pattern detector tier alongside it (per-model pii_detection.builtins / .patterns + core/services/routing/piipattern). - Detection labelled by source (ner vs pattern); backend trace / confidence / debug observability; analyze/redact exposed as a synchronous API. - Instance-wide default detector policy + per-usecase default-on; request filtering extended to completions, embeddings, edits & Ollama. - React UI: NER-centric PII editor, detector-models table, pattern/builtins editor, middleware default-policy UI. - Gallery: privacy-filter-multilingual token-classify model + NER install filter; token_classify known_usecase; batch sized to context for NER models. privacy-filter backend registered in the backend gallery (cpu/vulkan/cuda-13 meta + image entries with a capabilities map) matching its CI matrix jobs, and an /import-model auto-detect importer (PrivacyFilterImporter, narrow privacy-filter GGUF detection) replacing the prior pref-only registration. Reconciled against master's independent evolution: - Dropped master's PIIPatternOverrides feature (global-pattern runtime overrides + /api/pii/patterns API + runtime_settings.json persistence). The per-model NER + pattern-detector design supersedes it; it was built on the global redactor pattern set this branch replaced. - Reverted the llama.cpp Score carry-patch (0006-server-task-type-score): removed the patch and restored master's grpc-server.cpp Score RPC (direct llama_decode, slot-loop bypass) and LLAMA_VERSION pin, plus master's model_config validation forbidding score + chat/completion/embeddings on llama-cpp. token_classify is unaffected (it runs on the privacy-filter backend, not llama-cpp). Assisted-by: Claude:claude-opus-4-8 [Claude Code] Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-06-18 11:45:22 +01:00
LocalAI [bot]	294170d3ed	feat(backend): add depth-anything (Depth Anything 3) C++/ggml backend + gallery (#10352) ... * feat(backend): add depth-anything (Depth Anything 3) C++/ggml backend + gallery Mirrors the locate-anything-cpp backend to register a new depth-anything backend that wraps the Depth Anything 3 ggml port (depth-anything.cpp) via purego (cgo-less, no Python at inference). - backend/go/depth-anything-cpp/: gRPC backend (Load + Predict + GenerateImage), purego binding to the da_capi_* C ABI, CMake/Makefile/run/package/test scripts building depth-anything.cpp's DA_SHARED static .so per CPU variant. - backend/index.yaml: depth-anything backend meta + all hardware-variant capability entries (cpu/cuda12/cuda13/intel-sycl-f32+f16/vulkan/nvidia-l4t). - gallery/index.yaml: 8 Depth Anything 3 GGUF models (base q4_k/q8_0/f16/f32, small, large, giant, mono-large). - .github/backend-matrix.yml: one build entry per hardware variant. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(depth): typed Depth RPC + REST endpoint exposing full DA3 data Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(depth): pin depth-anything.cpp to e0b6814 (ABI 3 dense C-API) The Depth RPC handler calls da_capi_depth_dense / da_capi_points (C-API ABI 3); pin the native build to the commit that exports them. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(depth): pin depth-anything.cpp to v0.1.0 release (b515c31) Repoint the native version from the now-orphaned e0b6814 to the b515c31 release commit, kept alive by the upstream v0.1.0 tag. C-API is unchanged (da_capi_abi_version == 3). Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(depth): wire depth-anything-cpp into build, CI bump, and importer The backend dir, gallery index, and CI build-matrix were present but the backend was never wired into the integration points that adding-backends.md requires: - root Makefile: add to .NOTPARALLEL, the test-extra chain, a BACKEND_* definition, the docker-build target eval, and docker-build-backends (mirrors parakeet-cpp; the backend's own Makefile already documented that its `test` target is driven by test-extra). - bump_deps.yaml: register the DEPTHANYTHING_VERSION pin so the daily auto-bump bot tracks mudler/depth-anything.cpp master (it cannot see an unregistered Makefile pin). - import form: add a preference-only KnownBackend entry so depth-anything is selectable at /import-model (mirrors sam3-cpp; no reliable GGUF auto-detect signal, so pref-only per the doc's default). changed-backends.js needs no entry: the generic golang suffix branch already resolves backend/go/depth-anything-cpp/. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(depth): auto-detect importer for depth-anything GGUFs Replace the preference-only entry with a real auto-detect importer (mirrors parakeet-cpp / locate-anything): - DepthAnythingImporter matches a .gguf whose name carries a depth-anything token (depth-anything-<size>-<quant>.gguf), so /import-model recognises mudler/depth-anything.cpp-gguf repos and direct GGUF URLs without an explicit backend preference. preferences.backend= "depth-anything" still forces it. - Registered before LlamaCPPImporter so its GGUF bundles aren't claimed by the generic .gguf importer; the narrow name match means it cannot claim arbitrary llama GGUFs or the upstream safetensors PyTorch repos. - Multi-quant repos pick the smallest quant by default (q4_k -> ... -> f32, depth stays >0.998 corr even at q4_k); quantizations preference overrides. - Drops the now-redundant knownPrefOnlyBackends entry (importer-backed backends are not listed there, matching parakeet-cpp). - Table-driven Ginkgo test covers detection, negative cases (llama GGUF, upstream safetensors), default/override/fallback quant pick, and direct URL import. 10/10 specs pass. Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(depth): check conn.Close error in grpc Depth client (errcheck) The new Depth() client method used a bare `defer conn.Close()`. golangci-lint runs with new-from-merge-base, so although the 39 sibling methods use the same bare form (grandfathered), the newly added line trips errcheck. Drop the result explicitly to satisfy the linter. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-8 * fix(depth): bump depth-anything.cpp to v0.1.1 (embeddable CMake) v0.1.0 (b515c31) used ${CMAKE_SOURCE_DIR} for its include dirs, which points at the parent project when built via add_subdirectory() as this backend does, so the container build failed with missing stb_image.h / da_gguf_keys.h. v0.1.1 (2d42897) switches to project-relative paths. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-8 * fix(depth): resolve gosec findings in the backend wrapper The code-scanning gate flagged three new failure-level alerts in godepthanythingcpp.go (gosec runs with -no-fail; GitHub gates on new alerts): - G301: export dirs were created with 0o755. Tighten to 0o750 (no world access needed for backend-written export output). - G304: writeDepthPNG creates req.GetDst(). That path is chosen by the LocalAI core as the intended output destination (same pattern every image backend uses), not attacker input, so annotate with #nosec G304 and document why. The remaining G103 "audit unsafe" notes on the unsafe.Slice C-buffer copies are warning-level (the same purego interop whisper/parakeet use) and do not gate the check, per the supertonic exclusion precedent in secscan.yaml. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-8 * fix(depth): bump depth-anything.cpp to v0.1.2 (CUDA cross-build arch) v0.1.1 forced CMAKE_CUDA_ARCHITECTURES=native, which breaks the GPU-less l4t/cublas CI builds (nvcc "Unsupported gpu architecture 'compute_'" on CMake 3.22). v0.1.2 (442eea4) drops the override and lets ggml pick its default cross-build arch list. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-8 --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-06-16 16:28:28 +02:00
Richard Palethorpe	085fc53bbc	fix(router): production-ready request router + auto-size batch for embedding/rerank (#10104) ... * fix(router): score classifier production-readiness Conversation trimming runs through the classifier model's chat template and trims by exact token count, sized to the model's n_batch which is now scaled to context so long probes can't crash the backend. Missing chat_message templates are a hard error at router build time. Router- facing factories (Embedder/Scorer/Reranker/TokenCounter) re-resolve ModelConfig per call so a model installed post-startup doesn't bind a stub Backend="" config and silently fall into the loader's auto- iterate path. New 'vector_store' backend trace recorded inside localVectorStore on every Search/Insert — including the backend-load-failure path that previously vanished into an xlog.Warn — with outcome tagging (hit/miss/empty_store/backend_load_error/find_error/insert_error/ok). Companion cleanup drops misleading similarity:0 and input_tokens_count:0 from non-hit and text-mode traces. Gallery local-store-development aliases to 'local-store' so the master image satisfies pkg/model.LocalStoreBackend lookups from the embedding cache. Misc: llama-cpp TokenizeString reads the correct 'prompt' JSON key (the original bug); ModelTokenize nil-guard; non-fatal mitm proxy startup; PII 'route_local' renamed to 'allow' with docs/UI in sync; model-editor footer no longer eats the edit area on small screens; several config-editor template/dropdown/section fixes. Tests: e2e router specs (casual/code-hint + long-conversation trim), vector_store trace specs, lazy-factory specs, gallery dev-alias resolution, Playwright trace badge + scroll regression. Assisted-by: Claude:claude-opus-4-7 [Claude Code] Signed-off-by: Richard Palethorpe <io@richiejp.com> * feat(backend): auto-size batch to context for embedding and rerank models Embedding and rerank models pool over the whole input in a single physical batch (n_ubatch). With batch left at the 512 default, the backend rejects longer inputs with "input is too large to process", silently capping a large-context embedder (e.g. 8k/32k) at 512 tokens. Size n_batch to the context for these single-pass usecases, mirroring the existing FLAG_SCORE behaviour; an explicit batch: still wins. Extracts EffectiveContextSize/EffectiveBatchSize from grpcModelOpts so the effective decode window has one home for other callers to reuse. Adds an e2e-aio regression test that embeds a >512-token input. The AIO embedding model is switched to nomic-embed-text-v1.5 (2048 context) because the previous granite model was capped at 512 tokens and could not exercise the larger batch. Assisted-by: claude-code:claude-opus-4-8 [Claude Code] Signed-off-by: Richard Palethorpe <io@richiejp.com> * fix(gallery): raise arch-router scoring output cap via parallel:64 Scoring decodes the whole prompt+candidate in a single llama_decode and reads one logit row per candidate token. The vendored llama.cpp server caps causal output rows at n_parallel, so the default of 1 aborts with GGML_ASSERT(n_outputs_max <= cparams.n_outputs_max) on multi-token route labels. Set options: [parallel:64] on both arch-router quant entries to lift the cap; kv_unified (the grpc-server default) keeps the full context per sequence, so this does not split the KV cache. Assisted-by: claude-code:claude-opus-4-8 [Claude Code] Signed-off-by: Richard Palethorpe <io@richiejp.com> --------- Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-06-12 16:21:15 +02:00
LocalAI [bot]	27e63b9a78	feat(tts): support per-request instructions and params (#10172) ... The OpenAI-compatible TTS endpoint accepts an `instructions` field, but it was silently dropped at the HTTP->gRPC boundary: neither schema.TTSRequest nor the gRPC TTSRequest proto carried it, so backends could only read such a value from static YAML options (identical for every request). This blocked per-line emotion/style and, for Qwen3-TTS VoiceDesign, limited a model config to a single designed voice. Plumb a generic per-request instruction string end to end, plus an optional backend-specific params map: - proto: add `optional string instructions` and `map<string,string> params` to TTSRequest. - schema: add Instructions (maps OpenAI `instructions`) and Params (LocalAI extension) to schema.TTSRequest. - core: thread both through ModelTTS/ModelTTSStream via a newTTSRequest helper that attaches instructions only when non-empty (so backends can fall back to YAML when unset); forward them from the /v1/audio/speech handler. - qwen-tts: prefer the per-request instruction over the YAML `instruct` option (used by both mode detection and generation) and merge per-request params. - chatterbox: merge per-request params (coerced to float/int/bool) over YAML options into generate() kwargs. Fully backward compatible: empty instructions fall back to the YAML option and backends that don't support style/voice instructions ignore the field. Closes #10164 Assisted-by: Claude:claude-opus-4-8 [Claude Code] Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-06-04 11:45:02 +02:00
Tai An	0fd666ee6e	fix(openresponses): populate Content and accept bare {role,content} items (#10039) (#10040) ... * fix(openresponses): populate Content and accept bare {role,content} items (#10039) Fixes mudler/LocalAI#10039 — `/v1/responses` silently returned empty output on any model whose YAML doesn't include a Go-side `template.chat_message` block. Three cooperating bugs: * `convertORInputToMessages` populated only `StringContent` for string input and for the `input.Instructions` system message, leaving the `Content` (any) field nil. * `TemplateMessages` gated all fallback content-rendering branches on `Content != nil && StringContent != ""` — but every branch in that function consumes `StringContent`, not `Content`. The `&&` silently dropped messages that had StringContent set and Content nil, producing an empty prompt that the 5× empty-retry guard then turned into a 200 OK with `output: []`. * The array-input branch of `convertORInputToMessages` dispatched on `itemMap["type"]` with no default, dropping bare `{role, content}` items emitted by the OpenAI Python SDK helper `client.responses.create(input=[{...}])`. Fix: * Set both `Content` and `StringContent` in the two openresponses message-construction sites that only set one. * Treat a bare `{role, content}` item (no `type`) as `type: "message"` for OpenAI-SDK compatibility. * Gate `TemplateMessages` fallback rendering on `StringContent != ""`, which is what every downstream branch in that function actually reads. Regression test added to `evaluator_test.go` covering the fallback path (no `ChatMessage` template) with a StringContent-only message, both with and without a role mapping. * test(openresponses): guard Content population and ToProto path (#10039) Add regression tests for the two seams the original fix touched but left uncovered: * convertORInputToMessages must populate both Content and StringContent for plain string input and for bare {role, content} array items (the OpenAI SDK shape that omits the type discriminator). Both are functional reds against the pre-fix code. * Messages.ToProto reads Content, not StringContent — this is the path UseTokenizerTemplate backends (imported GGUFs) take. The cases pin that contract so a future regression on the producer side is caught. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-7 [Claude Code] --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-05-28 07:21:48 +00:00
Richard Palethorpe	6a80e23733	feat(middleware): Model routing, PII filtering, Cloud model proxies (#9802) ... Add a routing middleware stack and a cloud-proxy backend. * cloud-proxy: a Go gRPC backend that forwards OpenAI- and Anthropic-shaped chat requests to upstream providers, with an optional translate mode (OpenAI request -> Anthropic /v1/messages -> OpenAI response) and full tool-calling support. * routing: admission control, content-aware model routing (embedding cache + classifier + rerank + Arch-Router score), PII detection/redaction (regex + NER) with streaming filter and OpenAI/Anthropic adapters, and a per-user/per-key billing recorder backed by GORM or in-memory storage. * middleware: UsageMiddleware records usage via the billing recorder, plus admission, route-model, usage-stamp and trace middlewares. * observability: BackendTrace ring buffer stores full request bodies (capped), MITM proxy emits structured trace events, and router classifier decisions surface at /api/router/decide. * gallery: Arch-Router-1.5B (Q4_K_M and Q8_0). * UI: cloud-proxy model-editor fields, classifier system-prompt and score-normalization config, and a Traces page rendering request bodies. Assisted-by: claude-code:claude-opus-4-7 [Read] [Edit] [Bash] Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-05-25 09:28:27 +02:00
LocalAI [bot]	661a0c3b9d	fix(ollama): accept float-encoded integer options (fixes #9837) (#9849) ... fix(ollama): accept float-encoded integer options (num_ctx, top_k, ...) Home Assistant's Ollama integration encodes integer options as JSON floats (e.g. `"num_ctx": 8192.0`). Stdlib `json.Unmarshal` refuses to decode a number with fractional notation into an `int` field, so the entire request was rejected with HTTP 400 before reaching the backend: Unmarshal type error: expected=int, got=number 8192.0, field=options.num_ctx Add a custom `UnmarshalJSON` on `OllamaOptions` that routes the int fields (`top_k`, `num_predict`, `seed`, `repeat_last_n`, `num_ctx`) through `*json.Number`, then converts via `Int64()` with a `Float64()` fallback. Public field types are unchanged, so endpoint code is untouched. Float fields and `stop` continue to parse via the default path. Fixes #9837 Assisted-by: Claude Code:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-05-16 18:38:19 +02:00
LocalAI [bot]	8af963bdd9	fix(streaming): comply with OpenAI usage / stream_options spec (#9815) ... * fix(streaming): comply with OpenAI usage / stream_options spec (#8546) LocalAI emitted `"usage":{"prompt_tokens":0,...}` on every streamed chunk because `OpenAIResponse.Usage` was a value type without `omitempty`. The official OpenAI Node SDK and its consumers (continuedev/continue, Kilo Code, Roo Code, Zed, IntelliJ Continue) filter on a truthy `result.usage` to detect the trailing usage chunk; LocalAI's zero-but-non-null usage on every intermediate chunk made that filter swallow every content chunk and surface an empty chat response while the server log looked successful. Changes: - `core/schema/openai.go`: `Usage *OpenAIUsage \`json:"usage,omitempty"\`` so intermediate chunks no longer carry a `usage` key. Add `OpenAIRequest.StreamOptions` with `include_usage` to mirror OpenAI's request field. - `core/http/endpoints/openai/chat.go` and `completion.go`: keep using the `Usage` struct field as an in-process channel for the running cumulative, but strip it before JSON marshalling. When the request set `stream_options.include_usage: true`, emit a dedicated trailing chunk with `"choices": []` and the populated usage (matching the OpenAI spec and llama.cpp's server behavior). - `chat_emit.go`: new `streamUsageTrailerJSON` helper; drop the `usage` parameter from `buildNoActionFinalChunks` since chunks no longer carry usage. - Update `image.go`, `inpainting.go`, `edit.go` to wrap their Usage values with `&` for the new pointer field. - UI: send `stream_options:{include_usage:true}` from the React (`useChat.js`) and legacy (`static/chat.js`) chat clients so the token-count badge keeps populating now that the server is spec-compliant. Tests: - New `chat_stream_usage_test.go` pins the spec invariants: intermediate chunks have no `usage` key, the trailer JSON has `"choices":[]` and a populated `usage`, and `OpenAIRequest` parses `stream_options.include_usage`. - Update `chat_emit_test.go` to reflect that finals no longer embed usage. Verified against the live LocalAI instance: before the fix Continue's filter logic swallowed 16/16 token chunks; with the new shape it yields 4/5 and routes usage through the dedicated trailer chunk. Fixes #8546 Assisted-by: Claude:opus-4.7 [Claude Code] Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(streaming): silence errcheck on usage trailer Fprintf The new spec-compliant `stream_options.include_usage` trailer writes were flagged by errcheck since they're new code (golangci-lint runs new-from-merge-base on master); the surrounding `fmt.Fprintf` data: writes are grandfathered. Drop the return values explicitly to match the linter's contract without adding a nolint shim. Assisted-by: Claude:opus-4.7 [Claude Code] Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-05-14 08:53:46 +02:00
LocalAI [bot]	a57e73691d	fix(ollama): accept prompt alias on /api/embed for Ollama parity (#9780) ... Ollama's embedding endpoint accepts both `input` and `prompt` as the input string value (see ollama/ollama docs/api.md#generate-embeddings). LocalAI only accepted `input`, which broke client libraries that send the `prompt` form. Add `Prompt` to OllamaEmbedRequest and have GetInputStrings fall back to it when Input is unset. Input still wins when both are provided. Fixes #9767. Assisted-by: Claude:claude-opus-4-7 [Claude Code] Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-05-12 17:21:20 +02:00
LocalAI [bot]	bc3fb16105	feat(ollama): report model capabilities + details on /api/tags and /api/show (#9766) ... Ollama-compatible clients (Open WebUI, Enchanted, ollama-grid-search, etc.) rely on the `capabilities` list and `details.{parameter_size, quantization_level,families}` fields returned by /api/tags and /api/show to decide which models are eligible for a given task -- for example to filter the "embedding model" picker. Upstream Ollama returns these; LocalAI's compat layer was leaving them empty, so embedding models were silently rejected by clients that only allow chat models for chat and only allow embedding models for embeddings. This wires up the existing config signals already present in ModelConfig: - modelCapabilities() derives the Ollama capability strings from the config: "embedding" (FLAG_EMBEDDINGS), "completion" (FLAG_CHAT / FLAG_COMPLETION), "vision" (explicit KnownUsecases bit or MMProj / multimodal template / backend media marker), "tools" (auto-detected ToolFormatMarkers, JSON/Response regex, XML format, grammar triggers), "thinking" (ReasoningConfig with reasoning not disabled) and "insert" (presence of a completion template). - modelDetailsFromModelConfig() now fills families, parameter_size and quantization_level. The latter two are parsed from the GGUF filename via regex -- conservative tokens only (Q*/IQ*/F16/F32/BF16 and \d+(\.\d+)?[BM] surrounded by separators) so we don't accidentally match "Qwen3" as "3B". - modelInfoFromModelConfig() exposes general.architecture and general.context_length in the new ShowResponse.model_info map. Note: HasUsecases(FLAG_VISION) cannot be used directly -- GuessUsecases has no FLAG_VISION case and returns true at the end for any chat model. hasVisionSupport() instead reads KnownUsecases explicitly plus MMProj / template / media-marker signals. Tests are written first (TDD) using Ginkgo/Gomega -- DescribeTable for the capability mapping (embedding-only, chat, vision, thinking, tools via markers, tools via JSON regex, no-capability rerank) plus integration tests against ShowModelEndpoint that round-trip JSON through a real ModelConfigLoader populated from a temp YAML file. Fixes #9760. Assisted-by: Claude Code:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io>	2026-05-12 00:16:19 +02:00
Andreas Egli	af83518532	feat: support word-level timestamps for faster-whisper (#9621) ... Signed-off-by: Andreas Egli <github@kharan.ch> Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>	2026-05-06 00:32:52 +02:00
Ettore Di Giacinto	e86ade54a6	feat(api): add /v1/audio/diarization endpoint with sherpa-onnx + vibevoice.cpp (#9654) ... * feat(api): add /v1/audio/diarization endpoint with sherpa-onnx + vibevoice.cpp Closes #1648. OpenAI-style multipart endpoint that returns "who spoke when". Single endpoint instead of the issue's three-endpoint sketch (refactor /vad, /vad/embedding, /diarization) — the typical client wants one call, and embeddings can land later as a sibling without breaking this surface. Response shape borrows from Pyannote/Deepgram: segments carry a normalised SPEAKER_NN id (zero-padded, stable across the response) plus the raw backend label, optional per-segment text when the backend bundles ASR, and a speakers summary in verbose_json. response_format also accepts rttm so consumers can pipe straight into pyannote.metrics / dscore. Backends: * vibevoice-cpp — Diarize() reuses the existing vv_capi_asr pass. vibevoice's ASR prompt asks the model to emit [{Start,End,Speaker,Content}] natively, so diarization is a by-product of the same pass; include_text=true preserves the transcript per segment, otherwise we drop it. * sherpa-onnx — wraps the upstream SherpaOnnxOfflineSpeakerDiarization C API (pyannote segmentation + speaker-embedding extractor + fast clustering). libsherpa-shim grew config builders, a SetClustering wrapper for per-call num_clusters/threshold overrides, and a segment_at accessor (purego can't read field arrays out of SherpaOnnxOfflineSpeakerDiarizationSegment[] directly). Plumbing: new Diarize gRPC RPC + DiarizeRequest / DiarizeSegment / DiarizeResponse messages, threaded through interface.go, base, server, client, embed. Default Base impl returns unimplemented. Capability surfaces all updated: FLAG_DIARIZATION usecase, FeatureAudioDiarization permission (default-on), RouteFeatureRegistry entries for /v1/audio/diarization and /audio/diarization, audio instruction-def description widened, CAP_DIARIZATION JS symbol, swagger regenerated, /api/instructions discovery map updated. Tests: * core/backend: speaker-label normalisation (first-seen → SPEAKER_NN, per-speaker totals, nil-safety, fallback to backend NumSpeakers when no segments). * core/http/endpoints/openai: RTTM rendering (file-id basename, negative duration clamping, fallback id). * tests/e2e: mock-backend grew a deterministic Diarize that emits raw labels "5","2","5" so the e2e suite verifies SPEAKER_NN remapping, verbose_json speakers summary + transcript pass-through (gated by include_text), RTTM bytes content-type, and rejection of unknown response_format. mock-diarize model config registered with known_usecases=[FLAG_DIARIZATION] to bypass the backend-name guard. Docs: new features/audio-diarization.md (request/response, RTTM example, sherpa-onnx + vibevoice setup), cross-link from audio-to-text.md, entry in whats-new.md. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-7 [Claude Code] * fix(diarization): correct sherpa-onnx symbol name + lint cleanup CI failures on #9654: * sherpa-onnx-grpc-{tts,transcription} and sherpa-onnx-realtime panicked at backend startup with `undefined symbol: SherpaOnnxDestroyOfflineSpeakerDiarizationResult`. Upstream's actual symbol is SherpaOnnxOfflineSpeakerDiarizationDestroyResult (Destroy in the middle, not the prefix); the rest of the diarization surface follows the same naming pattern. The mismatched name made purego.RegisterLibFunc fail at dlopen time and crashed the gRPC server before the BeforeAll could probe Health, taking down every sherpa-onnx test job — not just the diarization-related ones. * golangci-lint flagged 5 errcheck violations on new defer cleanups (os.RemoveAll / Close / conn.Close); wrap each in a `defer func() { _ = X() }()` closure (matches the pattern other LocalAI files use for new code, since pre-existing bare defers are grandfathered in via new-from-merge-base). * golangci-lint also flagged forbidigo violations: the new diarization_test.go files used testing.T-style `t.Errorf` / `t.Fatalf`, which are forbidden by the project's coding-style policy (.agents/coding-style.md). Convert both files to Ginkgo/Gomega Describe/It with Expect(...) — they get picked up by the existing TestBackend / TestOpenAI suites, no new suite plumbing needed. * modernize linter: tightened the diarization segment loop to `for i := range int(numSegments)` (Go 1.22+ idiom). Verified locally: golangci-lint with new-from-merge-base=origin/master reports 0 issues across all touched packages, and the four mocked diarization e2e specs in tests/e2e/mock_backend_test.go still pass. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-7 [Claude Code] * fix(vibevoice-cpp): convert non-WAV input via ffmpeg + raise ASR token budget Confirmed end-to-end against a real LocalAI instance with vibevoice-asr-q4_k loaded and the multi-speaker MP3 sample at vibevoice.cpp/samples/2p_argument.mp3: both /v1/audio/transcriptions and /v1/audio/diarization now succeed and return correctly attributed speaker turns for the full clip. Two latent issues surfaced once the diarization endpoint actually exercised the backend with a non-trivial input: 1. vv_capi_asr only accepts WAV via load_wav_24k_mono. The previous code passed the uploaded path straight through, so anything that wasn't already a 24 kHz mono s16le WAV failed at the C side with rc=-8 and the very unhelpful "vv_capi_asr failed". prepareWavInput shells out to ffmpeg ("-ar 24000 -ac 1 -acodec pcm_s16le") in a per-call temp dir, matching the rate the model was trained on; both AudioTranscription and Diarize now route through it. This is the same shape sherpa-onnx uses (utils.AudioToWav), but vibevoice needs 24 kHz rather than 16 kHz so we don't reuse that helper. 2. The C ABI's max_new_tokens defaults to 256 when 0 is passed. That's fine for a five-second clip but not for anything past ~10 s — vibevoice stops mid-JSON, the parse fails, and the caller sees a hard error. Pass a much larger budget (16 384 ≈ ~9 minutes of speech at the model's ~30 tok/s rate); generation stops at EOS so this is a cap rather than a target. 3. As a defensive belt-and-braces, mirror AudioTranscription's existing "fall back to a single segment if the model emits non-JSON text" pattern in Diarize, so partial / unusual model output never produces a 500. This kept the endpoint usable while diagnosing (1) and (2), and is the right behaviour to keep. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-7 [Claude Code] * fix(vibevoice-cpp): pass valid WAVs through directly so ffmpeg is not required at runtime Spotted by tests-e2e-backend (1.25.x): the previous fix forced every incoming audio file through `ffmpeg -ar 24000 ...`, which meant the backend container — which does not ship ffmpeg — failed even for the existing happy path where the caller already uploads a WAV. The container-side error was: rpc error: code = Unknown desc = vibevoice-cpp: ffmpeg convert to 24k mono wav: exec: "ffmpeg": executable file not found in $PATH Reading vibevoice.cpp's audio_io.cpp, `load_wav_24k_mono` uses drwav and already accepts any PCM/IEEE-float WAV at any sample rate, downmixes multi-channel input to mono, and resamples to 24 kHz internally. So the only inputs that genuinely need an external converter are non-WAV formats (MP3, OGG, FLAC, ...). Detect WAVs by RIFF/WAVE magic at bytes 0..3 / 8..11 and pass them straight through with a no-op cleanup; everything else still goes through ffmpeg with the same 24 kHz mono s16le target. The result: * Container builds without ffmpeg keep working for WAV uploads (the e2e-backends fixture is jfk.wav at 16 kHz mono s16le). * MP3 and other non-WAV inputs still get the new ffmpeg conversion path so the diarization endpoint stays useful. * If the caller uploads a non-WAV but ffmpeg isn't on PATH, the surfaced error is still descriptive enough to act on. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-7 [Claude Code] * fix(ci): make gcc-14 install in Dockerfile.golang best-effort for jammy bases The LocalVQE PR (bb033b16) made `gcc-14 g++-14` an unconditional apt install in backend/Dockerfile.golang and pointed update-alternatives at them. That works on the default `BASE_IMAGE=ubuntu:24.04` (noble has gcc-14 in main), but every Go backend that builds on `nvcr.io/nvidia/l4t-jetpack:r36.4.0` — jammy under the hood — now fails at the apt step: E: Unable to locate package gcc-14 This blocked unrelated jobs: backend-jobs(*-nvidia-l4t-arm64-{stablediffusion-ggml, sam3-cpp, whisper, acestep-cpp, qwen3-tts-cpp, vibevoice-cpp}). LocalVQE itself is only matrix-built on ubuntu:24.04 (CPU + Vulkan), so it doesn't actually need gcc-14 anywhere else. Make the gcc-14 install conditional on the package being available in the configured apt repos. On noble: identical behaviour to today (gcc-14 installed, update-alternatives points at it). On jammy: skip the gcc-14 stanza entirely and let build-essential's default gcc take over, which is what the other Go backends compile with anyway. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Assisted-by: Claude:claude-opus-4-7 [Claude Code] --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-05-05 15:10:13 +02:00
Richard Palethorpe	bb033b16a9	feat: add LocalVQE backend and audio transformations UI (#9640) ... feat(audio-transform): add LocalVQE backend, bidi gRPC RPC, Studio UI Introduce a generic "audio transform" capability for any audio-in / audio-out operation (echo cancellation, noise suppression, dereverberation, voice conversion, etc.) and ship LocalVQE as the first backend implementation. Backend protocol: - Two new gRPC RPCs in backend.proto: unary AudioTransform for batch and bidirectional AudioTransformStream for low-latency frame-by-frame use. This is the first bidi stream in the proto; per-frame unary at LocalVQE's 16 ms hop would be RTT-bound. Wire it through pkg/grpc/{client,server, embed,interface,base} with paired-channel ergonomics. LocalVQE backend (backend/go/localvqe/): - Go-Purego wrapper around upstream liblocalvqe.so. CMake builds the upstream shared lib + its libggml-cpu-*.so runtime variants directly — no MODULE wrapper needed because LocalVQE handles CPU feature selection internally via GGML_BACKEND_DL. - Sets GGML_NTHREADS from opts.Threads (or runtime.NumCPU()-1) — without it LocalVQE runs single-threaded at ~1× realtime instead of the documented ~9.6×. - Reference-length policy: zero-pad short refs, truncate long ones (the trailing portion can't have leaked into a mic that wasn't recording). - Ginkgo test suite (9 always-on specs + 2 model-gated). HTTP layer: - POST /audio/transformations (alias /audio/transform): multipart batch endpoint, accepts audio + optional reference + params[*]=v form fields. Persists inputs alongside the output in GeneratedContentDir/audio so the React UI history can replay past (audio, reference, output) triples. - GET /audio/transformations/stream: WebSocket bidi, 16 ms PCM frames (interleaved stereo mic+ref in, mono out). JSON session.update envelope for config; constants hoisted in core/schema/audio_transform.go. - ffmpeg-based input normalisation to 16 kHz mono s16 WAV via the existing utils.AudioToWav (with passthrough fast-path), so the user can upload any format / rate without seeing the model's strict 16 kHz constraint. - BackendTraceAudioTransform integration so /api/backend-traces and the Traces UI light up with audio_snippet base64 and timing. - Routes registered under routes/localai.go (LocalAI extension; OpenAI has no /audio/transformations endpoint), traced via TraceMiddleware. Auth + capability + importer: - FLAG_AUDIO_TRANSFORM (model_config.go), FeatureAudioTransform (default-on, in APIFeatures), three RouteFeatureRegistry rows. - localvqe added to knownPrefOnlyBackends with modality "audio-transform". - Gallery entry localvqe-v1-1.3m (sha256-pinned, hosted on huggingface.co/LocalAI-io/LocalVQE). React UI: - New /app/transform page surfaced via a dedicated "Enhance" sidebar section (sibling of Tools / Biometrics) — the page is enhancement, not generation, so it lives outside Studio. Two AudioInput components (Upload + Record tabs, drag-drop, mic capture). - Echo-test button: records mic while playing the loaded reference through the speakers — the mic naturally picks up speaker bleed, giving a real (mic, ref) pair for AEC testing without leaving the UI. - Reusable WaveformPlayer (canvas peaks + click-to-seek + audio controls) and useAudioPeaks hook (shared module-scoped AudioContext to avoid hitting browser context limits with three players on one page); migrated TTS, Sound, Traces audio blocks to use it. - Past runs saved in localStorage via useMediaHistory('audio-transform') — the history entry stores all three URLs so clicking re-renders the full triple, not just the output. Build + e2e: - 11 matrix entries removed from .github/workflows/backend.yml (CUDA, ROCm, SYCL, Metal, L4T): upstream supports only CPU + Vulkan, so we ship those two and let GPU-class hardware route through Vulkan in the gallery capabilities map. - tests-localvqe-grpc-transform job in test-extra.yml (gated on detect-changes.outputs.localvqe). - New audio_transform capability + 4 specs in tests/e2e-backends. - Playwright spec suite in core/http/react-ui/e2e/audio-transform.spec.js (8 specs covering tabs, file upload, multipart shape, history, errors). Docs: - New docs/content/features/audio-transform.md covering the (audio, reference) mental model, batch + WebSocket wire formats, LocalVQE param keys, and a YAML config example. Cross-links from text-to-audio and audio-to-text feature pages. Assisted-by: Claude:claude-opus-4-7 [Bash Read Edit Write Agent TaskCreate] Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-05-04 22:07:11 +02:00
Ettore Di Giacinto	f5eb13d3c2	feat(insightface): add antispoofing (liveness) detection (#9515) ... * feat(insightface): add antispoofing (liveness) detection Light up the anti_spoofing flag that was parked during the first pass. Both FaceVerify and FaceAnalyze now run the Silent-Face MiniFASNetV2 + MiniFASNetV1SE ensemble (~4 MB, Apache 2.0, CPU <10ms) when the flag is set. Failed liveness on either image vetoes FaceVerify regardless of embedding similarity. Every insightface* gallery entry now ships the MiniFASNet ONNX weights so existing packs light up after reinstall. Setting the flag against a model without the MiniFASNet files returns FAILED_PRECONDITION (HTTP 412) with a clear install message — no silent is_real=false. FaceVerifyResponse gained per-image img{1,2}_is_real and img{1,2}_antispoof_score (proto 9-12); FaceAnalysis's existing is_real/antispoof_score fields are now populated. Schema fields are pointers so they are fully absent from the JSON response when anti_spoofing was not requested — avoids collapsing "not checked" with "checked and fake" under Go's omitempty on bool. Validated end-to-end over HTTP against a local install: - verify + anti_spoofing, both real -> verified=true, score ~0.76 - verify + anti_spoofing, img2 spoof -> verified=false, img2_is_real=false - analyze + anti_spoofing -> is_real and score per face - flag against model without MiniFASNet -> HTTP 412 fail-loud Assisted-by: Claude:claude-opus-4-7 go vet * test(insightface): wire test target into test-extra The root Makefile's `test-extra` already runs `$(MAKE) -C backend/python/insightface test`, but the backend's Makefile never defined the target — so the command silently errored and the suite was never executed in CI. Adding the two-line target (matching ace-step/Makefile) hooks `test.sh` → `runUnittests` → `python -m unittest test.py`, which discovers both the pre-existing engine classes (InsightFaceEngineTest, OnnxDirectEngineTest) and the new AntispoofingTest. Each class skips gracefully when its weights can't be downloaded from a network-restricted runner. Assisted-by: Claude:claude-opus-4-7 * test(insightface): exercise antispoofing in e2e-backends (both paths) Add a `face_antispoof` capability to the Ginkgo e2e suite and extend the existing FaceVerify + FaceAnalyze specs with liveness assertions covering BOTH paths: real fixture -> is_real=true, score>0, verified stays true spoof fixture -> is_real=false, verified vetoed to false The spoof fixture is upstream's own `image_F2.jpg` (via the yakhyo mirror) — verified locally against the MiniFASNetV2+V1SE ensemble to classify as is_real=false with score ~0.013. That makes the assertion deterministic across CI runs; synthetic/derived spoofs fool the model unpredictably and would be flaky. Makefile wires it up end-to-end: - New INSIGHTFACE_ANTISPOOF_* cache dir + two ONNX downloads with pinned SHAs, matching the gallery entries. - insightface-antispoof-models target shared by both backend configs. - FACE_SPOOF_IMAGE_URL passed via BACKEND_TEST_FACE_SPOOF_IMAGE_URL. - Both e2e targets (buffalo-sc + opencv) now: * depend on insightface-antispoof-models * pass antispoof_v2_onnx / antispoof_v1se_onnx in BACKEND_TEST_OPTIONS * include face_antispoof in BACKEND_TEST_CAPS backend_test.go adds the new capability constant and a faceSpoofFile fixture resolved the same way as faceFile1/2/3. Spoof assertions are gated on both capFaceAntispoof AND faceSpoofFile being set, so a test config that omits the spoof fixture degrades gracefully to "real path only" instead of failing. Assisted-by: Claude:claude-opus-4-7 go vet	2026-04-23 18:28:15 +02:00
Ettore Di Giacinto	181ebb6df4	feat: voice recognition (#9500) ... * feat(voice-recognition): add /v1/voice/{verify,analyze,embed} + speaker-recognition backend Audio analog to face recognition. Adds three gRPC RPCs (VoiceVerify / VoiceAnalyze / VoiceEmbed), their Go service and HTTP layers, a new FLAG_SPEAKER_RECOGNITION capability flag, and a Python backend scaffold under backend/python/speaker-recognition/ wrapping SpeechBrain ECAPA-TDNN with a parallel OnnxDirectEngine for WeSpeaker / 3D-Speaker ONNX exports. The kokoros Rust backend gets matching unimplemented trait stubs — tonic's async_trait has no defaults, so adding an RPC without Rust stubs breaks the build (same regression fixed by eb01c772 for face). Swagger, /api/instructions, and the auth RouteFeatureRegistry / APIFeatures list are updated so the endpoints surface everywhere a client or admin UI looks. Assisted-by: Claude:claude-opus-4-7 * feat(voice-recognition): add 1:N identify + register/forget endpoints Mirrors the face-recognition register/identify/forget surface. New package core/services/voicerecognition/ carries a Registry interface and a local-store-backed implementation (same in-memory vector-store plumbing facerecognition uses, separate instance so the embedding spaces stay isolated). Handlers under /v1/voice/{register,identify,forget} reuse backend.VoiceEmbed to compute the probe vector, then delegate the nearest-neighbour search to the registry. Default cosine-distance threshold is tuned for ECAPA-TDNN on VoxCeleb (0.25, EER ~1.9%). As with the face registry, the current backing is in-memory only — a pgvector implementation is a future constructor-level swap. Assisted-by: Claude:claude-opus-4-7 * feat(voice-recognition): gallery, docs, CI and e2e coverage - backend/index.yaml: speaker-recognition backend entry + CPU and CUDA-12 image variants (plus matching development variants). - gallery/index.yaml: speechbrain-ecapa-tdnn (default) and wespeaker-resnet34 model entries. The WeSpeaker SHA-256 is a deliberate placeholder — the HF URI must be curl'd and its hash filled in before the entry installs. - docs/content/features/voice-recognition.md: API reference + quickstart, mirrors the face-recognition docs. - React UI: CAP_SPEAKER_RECOGNITION flag export (consumers follow face's precedent — no dedicated tab yet). - tests/e2e-backends: voice_embed / voice_verify / voice_analyze specs. Helper resolveFaceFixture is reused as-is — the only thing face/voice share is "download a file into workDir", so no need for a new helper. - Makefile: docker-build-speaker-recognition + test-extra-backend- speaker-recognition-{ecapa,all} targets. Audio fixtures default to VCTK p225/p226 samples from HuggingFace. - CI: test-extra.yml grows a tests-speaker-recognition-grpc job mirroring insightface. backend.yml matrix gains CPU + CUDA-12 image build entries — scripts/changed-backends.js auto-picks these up. Assisted-by: Claude:claude-opus-4-7 * feat(voice-recognition): wire a working /v1/voice/analyze head Adds AnalysisHead: a lazy-loading age / gender / emotion inference wrapper that plugs into both SpeechBrainEngine and OnnxDirectEngine. Defaults to two open-licence HuggingFace checkpoints: - audeering/wav2vec2-large-robust-24-ft-age-gender (Apache 2.0) — age regression + 3-way gender (female / male / child). - superb/wav2vec2-base-superb-er (Apache 2.0) — 4-way emotion. Both are optional and degrade gracefully when transformers or the model can't be loaded — the engine raises NotImplementedError so the gRPC layer returns 501 instead of a generic 500. Emotion classes pass through from the model (neutral/happy/angry/sad on the default checkpoint); the e2e test now accepts any non-empty dominant gender so custom age_gender_model overrides don't fail it. Adds transformers to the backend's CPU and CUDA-12 requirements. Assisted-by: Claude:claude-opus-4-7 * fix(voice-recognition): pin real WeSpeaker ResNet34 ONNX SHA-256 Replaces the placeholder hash in gallery/index.yaml with the actual SHA-256 (7bb2f06e…) of the upstream Wespeaker/wespeaker-voxceleb-resnet34-LM ONNX at ~25MB. `local-ai models install wespeaker-resnet34` now succeeds. Assisted-by: Claude:claude-opus-4-7 * fix(voice-recognition): soundfile loader + honest analyze default Two issues surfaced on first end-to-end smoke with the actual backend image: 1. torchaudio.load in torchaudio 2.8+ requires the torchcodec package for audio decoding. Switch SpeechBrainEngine._load_waveform to the already-present soundfile (listed in requirements.txt) plus a numpy linear resample to 16kHz. Drops a heavy ffmpeg-linked dep and the codepath we never exercise (torchaudio's ffmpeg backend). 2. The AnalysisHead was defaulting to audeering/wav2vec2-large-robust- 24-ft-age-gender, but AutoModelForAudioClassification silently mangles that checkpoint — it reports the age head weights as UNEXPECTED and re-initialises the classifier head with random values, so the "gender" output is noise and there is no age output at all. Make age/gender opt-in instead (empty default; users wire a cleanly-loadable Wav2Vec2ForSequenceClassification checkpoint via age_gender_model: option). Emotion keeps its working Superb default. Also broaden _infer_age_gender's tensor-shape handling and catch runtime exceptions so a dodgy age/gender head never takes down the whole analyze call. Docs and README updated to match the new policy. Verified with the branch-scoped gallery on localhost: - voice/embed → 192-d ECAPA-TDNN vector - voice/verify → same-clip dist≈6e-08 verified=true; cross-speaker dist 0.76–0.99 verified=false (as expected) - voice/register/identify/forget → round-trip works, 404 on unknown id - voice/analyze → emotion populated, age/gender omitted (opt-in) Assisted-by: Claude:claude-opus-4-7 * fix(voice-recognition): real CI audio fixtures + fixture-agnostic verify spec Two issues surfaced after CI actually ran the speaker-recognition e2e target (I'd curl-tested against a running server but hadn't run the make target locally): 1. The default BACKEND_TEST_VOICE_AUDIO_* URLs pointed at huggingface.co/datasets/CSTR-Edinburgh/vctk paths that return 404 (the dataset is gated). Swap them for the speechbrain test samples served from github.com/speechbrain/speechbrain/raw/develop/ — public, no auth, correct 16kHz mono format. 2. The VoiceVerify spec required d(file1,file2) < 0.4, assuming file1/file2 were same-speaker. The speechbrain samples are three different speakers (example1/2/5), and there is no easy un-gated source of true same-speaker audio pairs (VoxCeleb/VCTK/LibriSpeech are all license- or size-gated for CI use). Replace the ceiling check with a relative-ordering assertion: d(pair) > d(same-clip) for both file2 and file3 — that's enough to prove the embeddings encode speaker info, and it works with any three non-identical clips. Actual speaker ordering d(1,2) vs d(1,3) is logged but not asserted. Local run: 4/4 voice specs pass (Health, LoadModel, VoiceEmbed, VoiceVerify) on the built backend image. 12 non-voice specs skipped as expected. Assisted-by: Claude:claude-opus-4-7 * fix(ci): checkout with submodules in the reusable backend_build workflow The kokoros Rust backend build fails with failed to read .../sources/Kokoros/kokoros/Cargo.toml: No such file because the reusable backend_build.yml workflow's actions/checkout step was missing `submodules: true`. Dockerfile.rust does `COPY . /LocalAI`, and without the submodule files the subsequent `cargo build` can't find the vendored Kokoros crate. The bug pre-dates this PR — scripts/changed-backends.js only triggers the kokoros image job when something under backend/rust/kokoros or the shared proto changes, so master had been coasting past it. The voice-recognition proto addition re-broke it. Other checkouts in backend.yml (llama-cpp-darwin) and test-extra.yml (insightface, kokoros, speaker-recognition) already pass `submodules: true`; this brings the shared backend image builder in line. Assisted-by: Claude:claude-opus-4-7	2026-04-23 12:07:14 +02:00
Ettore Di Giacinto	f0c92610a1	feat(importer): expand importer flow to almost all backends (#9466) ... * docs(agents): require importer integration when adding backends Document the importer registry workflow so contributors know that adding a new backend also requires updating the /import-model dropdown source: either a new importer in core/gallery/importers/, extending an existing one for drop-in replacements, or the pref-only slice for backends with no reliable auto-detect signal. Always covered by a table-driven test. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for Batch 0 primitives Introduce failing tests that drive Batch 0 of the importer expansion: - pkg/huggingface-api: assert GetModelDetails populates PipelineTag and LibraryName from /api/models/{repo}, and that a failing metadata endpoint still returns file details (best-effort fetch). - core/gallery/importers/helpers_test.go: new table-driven coverage for HasFile, HasExtension, HasONNX, HasONNXConfigPair, HasGGMLFile. - core/gallery/importers/importers_test.go: assert ErrAmbiguousImport sentinel exists and round-trips through errors.Is. - core/gallery/importers/local_test.go: extend with detection cases for ggml-*.bin (whisper), silero_vad.onnx (silero-vad), and the piper .onnx + .onnx.json pair. - core/http/endpoints/localai/import_model_test.go: assert ImportModelURIEndpoint returns HTTP 400 with a structured {error, detail, hint} body when ErrAmbiguousImport surfaces. All tests fail in the expected places (missing fields, missing helpers, missing sentinel, endpoint still wraps as 500). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): Batch 0 foundation — helpers, sentinel, local detection Implements the Batch 0 primitives that subsequent importer batches build on: - pkg/huggingface-api: ModelDetails gains PipelineTag and LibraryName. GetModelDetails now layers a best-effort GET /api/models/{repo} fetch on top of ListFiles — a metadata outage leaves the fields empty but still returns full file details. Uses a dedicated response struct because the single-model endpoint uses snake_case keys while the list endpoint historically returned camelCase. - core/gallery/importers/helpers.go: generic HasFile, HasExtension, HasONNX, HasONNXConfigPair, HasGGMLFile helpers working on []hfapi.ModelFile so per-backend importers can detect artefact patterns without duplicating string wrangling. - core/gallery/importers/importers.go: adds the ErrAmbiguousImport sentinel. DiscoverModelConfig now returns it (wrapped with fmt.Errorf("%w: ...")) when no importer matched AND the HF pipeline_tag falls in a whitelist of narrow modalities (ASR, TTS, sentence-similarity, text-classification, object-detection). The whitelist is intentionally narrow — unknown tags keep the previous "no importer matched" behaviour to avoid blocking rare repos. - core/gallery/importers/local.go: three new local-path detections, inserted before the existing merged-transformers branch: * ggml-*.bin → whisper * silero*.onnx → silero-vad * *.onnx + *.onnx.json pair → piper - core/http/endpoints/localai/import_model.go: ImportModelURIEndpoint surfaces ErrAmbiguousImport as HTTP 400 with {error, detail, hint} JSON, preserving existing behaviour for unrelated errors. Green tests: go test ./core/gallery/importers/... ./pkg/huggingface-api/... \ ./core/http/endpoints/localai/... Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(importers): red tests for KnownBackend endpoint and importer metadata Add failing tests that drive Batch UI-Dropdown: - importers_test.go: assert importers expose Name/Modality/AutoDetects and that LlamaCPPImporter advertises drop-in replacements via a new AdditionalBackendsProvider interface. A Registry() accessor is also expected. - backend_test.go (new): assert GET /backends/known returns []schema.KnownBackend, covers every importer, exposes drop-in llama-cpp replacements, includes curated pref-only backends, has no duplicates, and is sorted by Modality+Name. These tests fail at compile time against master; they are intentionally red so the follow-up green commit is reviewable. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery): add /backends/known endpoint for importer-aware backend list Extend the Importer interface with Name/Modality/AutoDetects so the import system can self-describe its registry, and introduce the AdditionalBackendsProvider interface so importers can advertise drop-in replacements (llama-cpp advertises ik-llama-cpp and turboquant). Expose the new GET /backends/known endpoint that merges: - the importer registry (auto-detect supported), - drop-in replacements hosted by importers (preference-only), - a curated knownPrefOnlyBackends slice for backends with no dedicated importer (sglang, tinygrad, trl, mlx-vlm, whisperx, kokoros, Qwen TTS variants, sam3-cpp) — kept at the top of backend.go so contributors adding a new pref-only backend have one obvious place to edit, - backends installed on disk but unknown to the importer (marked AutoDetect=false, empty Modality). The endpoint deliberately does NOT filter by gallery membership or host capability (unlike /backends/available): LocalAI may auto-install a backend that is not yet present, so the import form dropdown must show everything the importer knows about. Response is deduplicated (importer wins over pref-only) and sorted by Modality+Name for deterministic output. Registered in core/http/routes/localai.go next to /backends/available under the same admin middleware. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui): source import form backend dropdown from /backends/known Replace the hard-coded BACKENDS constant in ImportModel.jsx with a live fetch of /backends/known on mount. Users now see every backend the importer layer knows about (including preference-only entries) grouped by modality, not a stale subset. Changes: - config.js: add backendsKnown endpoint constant next to backendsAvailable. - api.js: add backendsApi.listKnown() wrapper. - ImportModel.jsx: remove BACKENDS constant, fetch the list via useEffect, and derive grouped options via buildBackendOptions. Preference-only entries render with a " (preference-only)" suffix. Loading state disables the dropdown with a "Loading backends…" placeholder; on fetch failure the form falls back to auto-detect only and surfaces a non-blocking toast. - SearchableSelect.jsx: accept items flagged isHeader=true and render them as non-selectable section dividers. Keyboard navigation skips headers and search queries hide them so filtered output stays relevant. Vitest is not set up in this project (devDependencies ship Playwright only). Per the brief's guard-rail, no frontend test framework is introduced; coverage is provided by the Go handler tests that assert the /backends/known contract consumed by the React form. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for whisper importer Asserts detection on ggerganov/whisper.cpp (via ggml-*.bin filename), the preferences.backend=whisper override path for arbitrary URIs, and the Importer interface metadata (name/modality/autodetect). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add whisper importer Recognises whisper.cpp GGML models by the "ggml-*.bin" filename convention (direct URL or HF repo member) and by the explicit preferences.backend="whisper" override. Emits backend: whisper with the transcript use-case. Registered before llama-cpp so the narrow filename signal wins before any generic GGUF match is attempted. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for moonshine importer Asserts detection on UsefulSensors/moonshine-tiny via owner + ONNX files, the preferences.backend=moonshine override for arbitrary URIs, and the Importer interface metadata (name/modality/autodetect). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add moonshine importer Matches UsefulSensors-owned HF repos whose artefacts or metadata identify them as ASR: on-disk .onnx files (the canonical Moonshine packaging) OR pipeline_tag=automatic-speech-recognition (covers transformers/safetensors-only sibling repos). preferences.backend= moonshine overrides detection. Test uses the live moonshine-tiny repo because the canonical UsefulSensors/moonshine repo currently hits a recursive-subfolder bug in pkg/huggingface-api ListFiles. Registered after WhisperImporter but before LlamaCPPImporter and TransformersImporter so the narrower owner+ASR signal wins before the generic tokenizer.json check routes the repo to transformers. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for nemo importer Asserts detection on nvidia/parakeet-tdt-0.6b-v3 via owner + .nemo file, the preferences.backend=nemo override for arbitrary URIs, and the Importer interface metadata (name/modality/autodetect). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add nemo importer Matches nvidia-owned HF repos that ship a .nemo checkpoint archive, the canonical NeMo ASR packaging. preferences.backend=nemo forces detection. Registered between moonshine and llama-cpp so the narrow owner + extension signal wins before any downstream generic matcher. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for faster-whisper importer Asserts detection on Systran/faster-whisper-large-v3 (owner + model.bin + config.json + ASR pipeline), the preferences.backend= faster-whisper override for arbitrary URIs, and the Importer interface metadata. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add faster-whisper importer Recognises CTranslate2-packaged whisper checkpoints distributed for the faster-whisper runtime: model.bin + config.json + ASR pipeline_tag, narrowed to Systran-owned repos or repo names containing "faster-whisper" to avoid falsely claiming vanilla OpenAI whisper HF repos. preferences.backend=faster-whisper overrides detection. Registered before llama-cpp and transformers so the narrow signal wins before tokenizer.json routes the repo to the generic transformers importer. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for qwen-asr importer Asserts detection on Qwen/Qwen3-ASR-1.7B via owner + ASR substring in the repo name, the preferences.backend=qwen-asr override for arbitrary URIs, and the Importer interface metadata. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add qwen-asr importer Matches Qwen-owned HF repos whose name contains "ASR" (case-insensitive), routing them to the qwen-asr backend rather than the generic transformers/vllm path. The substring check scans the repo portion only so the owner field cannot leak a false match. preferences.backend=qwen-asr forces detection. Registered before llama-cpp and transformers so the narrow owner+name signal wins. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): ASR ambiguity surfaces ErrAmbiguousImport Locks in the behaviour added in Batch 0: an HF repo whose pipeline_tag marks it as automatic-speech-recognition but whose artefacts match no ASR importer (and no generic importer) must fail with ErrAmbiguousImport so callers know to pass preferences.backend rather than silently guess. pyannote/voice-activity-detection is the fixture — its file list is only config.yaml + README, leaving every importer's artefact check negative. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for piper importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add piper importer Detects piper TTS voices by the canonical <voice>.onnx + <voice>.onnx.json pair packaging (via HasONNXConfigPair). Narrow enough to skip generic ONNX repos used by other backends (Moonshine ASR, sentence-transformers). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for bark importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add bark importer Detects Suno's Bark TTS checkpoints by HF owner "suno" + repo name prefix "bark". Adds HFOwnerRepoFromURI() helper so importers can fall back to URI parsing when pkg/huggingface-api's recursive tree listing errors on repos with nested subdirectories (suno/bark ships a speaker_embeddings/v2 subtree that trips a pre-existing path-doubling bug in the listFilesInPath recursion). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for fish-speech importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add fish-speech importer Detects Fish Audio TTS releases by HF owner "fishaudio" with a URI-based fallback for repos whose tree recursion trips the pre-existing hfapi path-doubling bug. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for outetts importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add outetts importer Detects OuteAI's OuteTTS releases by HF owner "OuteAI" or a case- insensitive "OuteTTS" substring in the repo name, with a URI-based fallback for recursion-bugged repos. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for voxcpm importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add voxcpm importer Detects OpenBMB's VoxCPM TTS family by repo-name substring (community mirrors re-host the weights under many owners — mlx-community, bluryar, callgg, etc). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for kokoro importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add kokoro importer Detects hexgrad's Kokoro TTS by the "Kokoro" repo-name substring paired with a PyTorch .pth/.pt checkpoint — the pairing excludes ONNX-only mirrors (handled by the pref-only `kokoros` Rust runtime) and GGUF mirrors (handled by llama-cpp). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for kitten-tts importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add kitten-tts importer Detects KittenML's kitten-tts releases by owner or "kitten-tts" repo-name substring, with URI-parsing fallback. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for neutts importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add neutts importer Detects Neuphonic's NeuTTS releases by owner "neuphonic" or "neutts" repo-name substring, with URI-parsing fallback. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for chatterbox importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add chatterbox importer Detects Resemble AI's Chatterbox TTS by owner "ResembleAI" or "chatterbox" repo-name substring, with URI-parsing fallback. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for vibevoice importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add vibevoice importer Detects Microsoft's VibeVoice TTS by "vibevoice" repo-name substring (case-insensitive) so community mirrors still route here. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for coqui importer Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add coqui importer Detects Coqui AI's TTS releases (XTTS-v2, YourTTS, …) by the authoritative `coqui` HF owner, with URI-parsing fallback. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): TTS ambiguity surfaces ErrAmbiguousImport Adds a Ginkgo spec that imports nari-labs/Dia-1.6B — a real HF repo carrying pipeline_tag="text-to-speech" whose artefacts (*.pth, one safetensors shard, preprocessor_config.json, config.json) match none of the Batch-2 TTS importers nor the generic text/image importers — and asserts DiscoverModelConfig wraps ErrAmbiguousImport via errors.Is. Also pivots the endpoint-level ambiguity fixture from hexgrad/Kokoro-82M to nari-labs/Dia-1.6B. Batch 2 added a dedicated kokoro importer that now claims the original fixture; Dia remains genuinely unclaimed and so exercises the same ambiguity code path at the HTTP layer. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for stablediffusion-ggml importer Covers HF repo detection (city96/FLUX.1-dev-gguf), raw .gguf URL matching on filename arch tokens, preference override, and Importer interface metadata. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add stablediffusion-ggml importer Detects GGUF-packed Stable Diffusion and FLUX checkpoints (leejet owner, city96 FLUX mirrors, second-state SD dumps, raw .gguf URLs with arch tokens) and routes them to the stablediffusion-ggml backend. Registered BEFORE LlamaCPPImporter so .gguf image checkpoints are not stolen by llama-cpp's generic .gguf match. Reuses HFOwnerRepoFromURI for the hfapi-recursion-bug fallback. preferences.backend overrides detection. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for ace-step importer Covers HF repo-name detection (ACE-Step/ACE-Step-v1-3.5B), preference override, and Importer interface metadata. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add ace-step importer Routes ACE-Step music generation checkpoints (ACE-Step/ACE-Step-v1-3.5B, ACE-Step/Ace-Step1.5, community mirrors) to the ace-step backend. Matching is case-insensitive on the "ace-step" repo-name substring and owner, with an HFOwnerRepoFromURI fallback for the hfapi recursion bug. KnownUsecaseStrings mirrors the gallery's ace-step-turbo entry (sound_generation, tts). preferences.backend overrides. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): surface ErrAmbiguousImport on text-to-image misses Adds text-to-image to ambiguousModalities whitelist and covers the h94/IP-Adapter-FaceID case — pipeline_tag=text-to-image but ships only .bin/.safetensors so diffusers, stablediffusion-ggml, llama-cpp, transformers, vllm, mlx, and ace-step all miss. DiscoverModelConfig now surfaces ErrAmbiguousImport for that shape instead of the opaque "no importer matched" error. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for vllm-omni importer Introduces the test surface for the forthcoming VLLMOmniImporter: detection via preferences.backend, Qwen owner + Omni repo token, URI-only fallback, negative cases (plain Qwen, random OmniX repo), and Import() emitting backend: vllm-omni with chat + multimodal usecases. Includes a registration-order assertion via DiscoverModelConfig to pin the requirement that vllm-omni wins over vllm for Qwen Omni repos (tokenizer files are usually present too). Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add vllm-omni importer Adds VLLMOmniImporter for Qwen Omni-style multimodal checkpoints (Qwen3-Omni, Qwen2.5-Omni, …). Detection is narrow: HF owner "Qwen" combined with "omni" in the repo name, or a repo name matching the -Omni-/Omni- naming pattern. preferences.backend="vllm-omni" always wins; HFOwnerRepoFromURI provides a URI-only fallback for the hfapi recursion-bug edge case. Emitted YAML sets backend: vllm-omni and known_usecases: [chat, multimodal], matching the gallery/index.yaml vllm-omni entries. The importer is registered ahead of VLLMImporter so Qwen Omni repos — which also carry tokenizer files — route to vllm-omni rather than the plain vllm backend. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for llama-cpp drop-in preferences Pins the expected drop-in replacement behaviour: preferences.backend of ik-llama-cpp or turboquant must swap the emitted YAML backend field while keeping the llama-cpp file layout identical. Also covers the unknown-backend case (must stay llama-cpp) and re-asserts AdditionalBackends() returns the two curated entries with non-empty descriptions. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): llama-cpp honours ik-llama-cpp and turboquant drop-in preferences preferences.backend set to ik-llama-cpp or turboquant now swaps the emitted YAML backend field while leaving the file layout, model path, mmproj handling and everything else in the llama-cpp Import pipeline untouched. Unknown values are ignored and fall back to backend: llama-cpp so arbitrary input can't leak into the config. Aligns the AdditionalBackends() descriptions with the user-facing naming conventions surfaced via /backends/known. No changes to the pref-only curated list in endpoints/localai/backend.go: the two drop-in names have always lived on the importer side via AdditionalBackends. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for silero-vad importer Add the SileroVADImporter test fixtures covering metadata, preference overrides, snakers4 + onnx detection, silero_vad.onnx canonical filename, URI fallback, and live HF discovery. Implementation follows in the next commit. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add silero-vad importer Recognise the Silero VAD ONNX packaging: the canonical silero_vad.onnx filename or any ONNX file under the snakers4 owner. Emits a backend: silero-vad config with the vad known_usecase, and attaches the canonical file entry when present so the weights download on import. Registered before the generic importers so the unique-filename signal takes precedence over any downstream tokenizer-based matcher. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for rerankers importer Cover the RerankersImporter contract: interface metadata, preference override, cross-encoder owner detection, case-insensitive 'reranker' substring match (BAAI/bge-reranker, Alibaba-NLP/gte-reranker), URI fallback, and the full-discovery ordering check that a BAAI reranker repo must route to the rerankers importer rather than transformers. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add rerankers importer Recognise reranker repositories — cross-encoder owner or any repo whose name contains 'reranker' (case-insensitive). Emits backend: rerankers with reranking: true and the rerank known_usecase. Registered ahead of sentencetransformers and transformers so reranker repos that happen to ship tokenizer.json or modules.json still route here. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for sentencetransformers importer Cover the SentenceTransformersImporter contract: interface metadata, preference override, modules.json marker file, sentence_bert_config.json marker file, sentence-transformers owner, URI fallback, and the full-discovery ordering check that ensures a sentence-transformers HF URI routes here rather than transformers. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add sentencetransformers importer Recognise sentence-transformers embedding repos by modules.json, sentence_bert_config.json, or the sentence-transformers owner. Emits backend: sentencetransformers with embeddings: true and the embeddings known_usecase. Registered ahead of transformers so ST repos that carry tokenizer.json still route here. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): add failing tests for rfdetr importer Cover the RFDetrImporter contract: interface metadata, preference override, case-insensitive rf-detr and rfdetr substring matches, URI fallback, and negative cases. Implementation follows in the next commit. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(gallery/importers): add rfdetr importer Recognise RF-DETR object-detection repositories by a case-insensitive 'rf-detr' / 'rfdetr' substring in the repo name. Emits backend: rfdetr with the detection known_usecase. Registered ahead of transformers so RF-DETR repos with tokenizer artefacts still route here. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(gallery/importers): surface ErrAmbiguousImport on sentence-similarity misses Add an ambiguity fixture covering the embeddings/rerankers modality. Qdrant/bm25 carries pipeline_tag=sentence-similarity but ships only config.json + stopword .txt files — none of the Batch 5 importers (silero-vad, rerankers, sentencetransformers, rfdetr) or the generic vllm/transformers/llama-cpp/mlx/diffusers importers match. Because the modality is in the ambiguous whitelist, DiscoverModelConfig must surface ErrAmbiguousImport. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(localai/backend): red tests for KnownBackend.Installed flag Extend the /backends/known suite with three failing cases that pin down the forthcoming Installed field: JSON field presence on every entry, flipping to true when an importer-registered backend is also present on disk (and staying false for non-installed pref-only entries), and surfacing system-only backends with empty modality and AutoDetect=false. A small writeFakeSystemBackend helper plants a run.sh under the backends dir so gallery.ListSystemBackends recognises the fixture. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(schema,localai/backend): add Installed flag to KnownBackend Add an Installed bool to schema.KnownBackend and populate it from the /backends/known handler so the React import form can warn users that picking a not-yet-installed backend will trigger an automatic download on submit. Computation: after merging the importer registry, additional backends provider entries and the curated pref-only slice, the handler walks gallery.ListSystemBackends(systemState) and either flips the existing map entry's Installed flag to true (preserving modality / autodetect / description metadata) or inserts a bare {Installed:true} entry for system-only backends the importer layer doesn't know about. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(localai/import_model): structured ambiguous-import response Add red tests covering the extended ambiguity shape the React import form needs: - ImportModelURIEndpoint must return an HTTP 400 body that exposes the detected `modality` (normalised to the importer modality key, e.g. "tts" for pipeline_tag=text-to-speech) and a list of `candidates` (backend names filtered by modality, excluding text-LLM backends). - The importers package must surface a typed AmbiguousImportError so HTTP consumers can read Modality + Candidates without parsing the error string. errors.Is against the existing sentinel keeps working. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(localai/import_model): structured ambiguity response with modality + candidates DiscoverModelConfig now returns a typed AmbiguousImportError that carries the importer modality key, candidate backend names, the original URI, and the raw HF pipeline_tag. Its Is() preserves errors.Is(err, ErrAmbiguousImport) for legacy callers. The importer modality is pre-mapped from the HF pipeline_tag (automatic-speech-recognition → asr, text-to-speech → tts, etc) via PipelineTagToModality — surfaced as an exported helper so downstream consumers can avoid duplicating the table. CandidatesForModality filters the default importer registry plus AdditionalBackendsProvider drop-ins by modality, sorts deterministically, and is the single source of truth used by ImportModelURIEndpoint. ImportModelURIEndpoint now returns HTTP 400 with { error, detail, modality, candidates, hint } when ambiguity fires, letting the React form render a modality-scoped picker inline instead of a generic toast. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): manual pick badge + tooltip Red Playwright coverage for the preference-only → manual pick rename: - The Backend dropdown renders a "manual pick" badge on every option whose KnownBackend.auto_detect is false. - The badge carries a title attribute with hover-tooltip copy that explains auto-detect won't route to this backend. - Auto-detectable backends must NOT carry the badge. - The legacy " (preference-only)" suffix is gone from every label. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * ui(import): replace preference-only suffix with manual pick badge SearchableSelect option rows now support an optional badge field — a muted pill rendered to the right of the label with an optional title attribute for native hover tooltips. Plain text so screen readers read it alongside the option name. buildBackendOptions in ImportModel stops appending " (preference-only)" to the label and instead sets badge="manual pick" plus a descriptive tooltip on every option whose auto_detect is false. The Backend help text explains what "manual pick" means so users aren't left wondering about the badge. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): inline ambiguity picker Red Playwright coverage for Batch A2 — when the server returns a 400 ambiguity body, the form must render an inline alert instead of a toast, expose one clickable chip per candidate backend, and support both auto-resubmit on pick and silent dismiss. - Mocks /api/models/import-uri with the structured ambiguity body (error, detail, modality, candidates, hint). - On first click of Import, the alert is visible, carries modality-specific copy, and shows a chip per candidate. - Clicking a chip clears the alert, sets the Backend dropdown, and triggers a second POST to /api/models/import-uri. - Dismissing the alert leaves the Backend dropdown on Auto-detect — no implicit backend assignment. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): inline ambiguity alert with candidate chips Adds AmbiguityAlert — a soft, info-coloured card rendered above the URI input when the server returns a structured 400 with { modality, candidates }. Message is modality-aware (tts/asr/embeddings/image/ reranker/detection get purpose-written copy, everything else falls back to a generic template). Each candidate is a clickable chip that shows a download icon when /backends/known marks the backend as not yet installed, so users aren't surprised by an implicit install. ImportModel wires the alert to handleSimpleImport's error path: - api.handleResponse now attaches { status, body } to the thrown Error so pages can pattern-match on structured responses instead of string error messages. - handleSimpleImport detects `status === 400 && body.error === 'ambiguous import'` and flips into the inline-picker mode instead of toasting. - Clicking a chip sets prefs.backend and auto-resubmits (passing the picked backend as an override so setPrefs's asynchrony doesn't leak a stale value). - Dismissing clears the alert; changing the URI or the backend also clears it so a stale alert never sticks around. Test fixtures mock GET /backends/known + POST /models/import-uri so the Playwright specs don't depend on real network reachability. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): auto-install warning Red Playwright coverage for Batch A3 — when the user picks a backend whose KnownBackend.installed is false, the form must render a muted inline note under the Backend dropdown warning that submitting will download the backend first. Picking an installed backend or leaving Auto-detect selected must keep the note hidden. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): auto-install warning under backend dropdown When the user picks a backend whose KnownBackend.installed is false, render a muted inline note under the Backend dropdown's help text warning that submitting will download the backend first. The note lives inside the same form-group so it lines up with the existing hint text; it's hidden when Auto-detect is selected (the selected backend is unknowable at that point) or when the chosen backend is already on disk. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * ui(import): drop redundant section header, adjust icons, rename HF shortcut - Remove the "Import from URI" card-level <h2> — the page title already says "Import New Model" one row up, so the secondary header was duplicating information. - Swap the fa-star on "Common Preferences" for fa-sliders (stars imply favourites/ratings; this is just a preferences block) and move the Custom Preferences fa-sliders-h to fa-plus-circle so the two blocks read as distinct rather than as two sliders. - Rename the HF shortcut from "Search GGUF on HF" → "Browse models on HF" and drop the `search=gguf` filter on the linked URL. The import form now supports ~40 backends; hard-coding GGUF in the copy no longer matches the form's actual reach. - Pure polish — no behaviour change, covered by the existing Batch A Playwright suite. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): batch B — simple/power switch, options, tabs, dialog Adds a failing Playwright suite covering the full Batch B surface ahead of implementation: - B1: SimplePowerSwitch segmented control renders, toggles, persists to localStorage across reloads. - B2: Simple-mode Options disclosure is collapsed by default; expanding exposes only Backend, Model Name, Description (no quantizations, mmproj, model type, or custom prefs). - B3: Power mode has Preferences and YAML tabs with a persistent selection across reloads; URI/name/description typed in Simple carry over to Power; YAML tab swaps the primary action to Create. - B4: Switching Power -> Simple with a custom preference set triggers the 3-button confirmation dialog (Keep / Discard / Cancel) with the documented semantics. Tests fail against master — implementation lands in the following commits. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): add SimplePowerSwitch segmented control Replaces the previous "Advanced Mode / Simple Mode" toggle button in the page header with a two-segment control that flips between Simple and Power. The control reuses the existing .segmented CSS shared with the Sound page for visual consistency. Mode state is persisted to localStorage under `import-form-mode` so reloads land on the same view (default: simple). The boolean alias `isAdvancedMode` is retained internally to minimise diff — subsequent commits reshape the Simple and Power surfaces independently. Closes B1 from the Batch B Playwright suite. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): simple mode collapsible options, power tabs, switch dialog Completes the Batch B surface in a single structural pass so Simple and Power mode can evolve independently: Simple mode - URI input + Ambiguity alert + Import button, plus a collapsible "Options" disclosure that exposes ONLY Backend, Model Name, Description. Quantizations / MMProj / Model Type / Diffusers fields / Custom Preferences are no longer rendered in Simple mode. Power mode - In-page segmented "Preferences · YAML" tab strip. Active tab persists to localStorage under `import-form-power-tab`. - Preferences tab = the full existing preferences + custom prefs panel (no progressive disclosure yet — that's Batch D). - YAML tab = the existing CodeEditor. Primary button reads "Create" here, "Import Model" everywhere else. Switch dialog - Power -> Simple with non-default prefs (advanced pref keys set, any custom-pref key non-empty, or YAML edited away from the template) opens a 3-button dialog: Keep & switch / Discard & switch / Cancel. - Keep preserves all state. Discard resets prefs + customPrefs + YAML to defaults. Cancel leaves the user in Power mode. Page subtitle reflects the current surface (Simple, Power/Preferences, Power/YAML). Estimate banner renders everywhere except Power/YAML. Closes B2/B3/B4 from the Batch B Playwright suite. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): expand Options disclosure in Batch A tests Batch B hid the Backend dropdown behind a collapsible Options disclosure in Simple mode. The Batch A tests that exercise the dropdown directly (manual-pick badge, ambiguity chip sets the selected backend, auto- install warning) now click the disclosure toggle before asserting on dropdown contents. Test intent is unchanged. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * ui(import): strip decorative icons from field labels The preference panel had 12 Font Awesome icons decorating field labels (Backend, Model Name, Description, Quantizations, MMProj Quantizations, Model Type, Pipeline Type, Scheduler Type, Enable Parameters, Embeddings, CUDA, plus fa-link on Model URI). Every label screamed equally, flattening the visual hierarchy. Remove them. Keep icons where they carry meaning: page-level section headers, URI format guide entries, primary buttons, the Simple-mode Options disclosure, the ambiguity alert's fa-lightbulb, the auto-install note's fa-download, and the Estimated-requirements banner's fa-memory / fa-microchip / fa-download. No new behaviour, no layout / spacing changes beyond removing the orphaned icon margin. Playwright suite green. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): progressive disclosure of preference fields Cover the Batch D visibility matrix for Power > Preferences: Quantizations, MMProj Quantizations, and Model Type each render only for the backends that can consume them, stay visible when the backend is unset, and preserve any value the user already typed when toggled off and back on. Also pin the shrunk Description textarea at rows=2. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): progressive disclosure + shorter description textarea Gate Quantizations, MMProj Quantizations, and Model Type in the Power > Preferences tab so each field only renders for the backends that can actually consume it. Backend unset keeps everything visible. Hidden fields' state is preserved (the JSX wrapper is guarded, not the underlying prefs state) so users flipping backends back and forth don't lose input. Also shrink the Description textarea from rows=3 to rows=2 — it's shared between Simple Options and Power Preferences so the change applies to both. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): enter-to-submit in Simple mode Red test for Batch F3 — pressing Enter in the URI input must POST /models/import-uri, and Enter in the Description textarea must insert a newline without submitting the form. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): enter-to-submit in Simple mode Wrap the Simple-mode URI input + ambiguity alert + Options disclosure in a <form> whose onSubmit calls handleSimpleImport. Pressing Enter in the URI input (or any Simple-mode text input) now submits the import without having to move the mouse to the header button. The Description textarea keeps its native behaviour — Enter inserts a newline. A hidden submit button is included because the visible Import button lives outside the form in the page header; some browsers only fire implicit Enter-submit when the form contains a submit-capable element. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * ui(import,SearchableSelect,components): aria-hidden on decorative icons Every Font Awesome icon in the import form is decorative — its meaning is already conveyed by adjacent visible text. Adding aria-hidden="true" prevents screen readers from announcing the unicode glyph point as content. Covers ImportModel.jsx (all remaining <i> glyphs) and SearchableSelect.jsx (the trigger chevron). AmbiguityAlert and SimplePowerSwitch already set aria-hidden on their icons when the components landed in Batches A and B — no change needed there. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * ui(SearchableSelect): responsive dropdown maxHeight + hover focus guard F2 — replace fixed pixel heights with min(pixel, vh) so the dropdown and its inner scroll region don't overflow short viewports. Outer container: 260px -> min(260px, 60vh); inner listbox: 200px -> min(200px, 50vh). Tall viewports still get the original pixel caps. F5 — short-circuit onMouseEnter when the hovered row is already the focused row. Avoids queueing a setFocusIndex call (and a render) for every mousemove inside the same item — the state would be identical. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * ui(import): aria-label on custom preference rows The Key / Value inputs and trash button in each Custom Preferences row previously relied on placeholder text alone. Placeholders are not accessible names — they vanish on input and screen readers do not announce them consistently. Add row-indexed aria-labels so assistive tech can distinguish "Preference key for row 1" from "row 2", and give the trash button an explicit "Remove this preference" label. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * test(ui/import): modality chip row Red tests for Batch E — a horizontal modality chip row that filters the Backend dropdown by modality. Covers visibility in Simple-mode Options and Power/Preferences (and absence in Power/YAML), filter behaviour, mismatched-backend clearing with toast, ambiguity-alert auto-selection, and radiogroup keyboard navigation. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * feat(ui/import): add ModalityChips component + filter integration Horizontal chip row (Any, Text, Speech, TTS, Image, Embeddings, Rerankers, Detection, VAD) filters the Backend dropdown options to the selected modality. Default is Any — no filter, current behaviour. - New ModalityChips component (radiogroup pattern, roving tabindex, arrow-key navigation, Home/End). - buildBackendOptions now accepts an optional modalityFilter so grouped output is narrowed before rendering. - Chips render inside Simple-mode Options disclosure and Power > Preferences tab. Power > YAML stays unaffected. - Switching the filter drops a mismatched backend selection and surfaces a toast so the auto-clear is visible. - Ambiguity alerts auto-activate the matching chip so users see only relevant backends even if they dismiss the alert. Tightens the Batch E tests' option-matching to the label <span> so the "↵" keybind hint on the focused row doesn't break accessible-name lookups. Assisted-by: Claude:claude-opus-4-7[1m] [Agent] * fix(ui/import): rename Power to Advanced + stop URI-formats toggle from submitting form The "Supported URI Formats" disclosure button inside the Simple-mode form lacked an explicit type attribute, so it defaulted to type="submit". Every click triggered the form's onSubmit and surfaced the empty-URI validation toast ("Please enter a model URI"). Marking it type="button" lets it behave as a pure toggle. While here, rename the user-visible "Power" label to "Advanced" in the mode switch (button text + tooltip) and the Power-mode tab's aria-label, matching the term users actually expect. The internal mode key stays 'power' so tests, localStorage, and data-testid selectors are untouched. Assisted-by: Claude:claude-opus-4-7 * fix(system): fall back to cpu when meta backend lacks default capability Meta backends like vllm and sglang enumerate concrete variants for nvidia/amd/intel/cpu but omit a default: catch-all entry. On a no-GPU host the reported capability is "default", so the previous Capability() returned "default" unconditionally on a miss — IsCompatibleWith then saw no "default" key and filtered the meta out of AvailableBackends. The import flow's auto-install step then failed with "no backend found with name <meta>", contradicting the UI's promise that the backend would be downloaded on demand. Try the explicit "default" key first, then fall back to "cpu" before giving up. vllm now resolves to cpu-vllm on CPU-only Linux without touching the gallery YAML. Assisted-by: Claude:claude-opus-4-7	2026-04-22 22:42:37 +02:00
Ettore Di Giacinto	20baec77ab	feat(face-recognition): add insightface/onnx backend for 1:1 verify, 1:N identify, embedding, detection, analysis (#9480) ... * feat(face-recognition): add insightface backend for 1:1 verify, 1:N identify, embedding, detection, analysis Adds face recognition as a new first-class capability in LocalAI via the `insightface` Python backend, with a pluggable two-engine design so non-commercial (insightface model packs) and commercial-safe (OpenCV Zoo YuNet + SFace) models share the same gRPC/HTTP surface. New gRPC RPCs (backend/backend.proto): * FaceVerify(FaceVerifyRequest) returns FaceVerifyResponse * FaceAnalyze(FaceAnalyzeRequest) returns FaceAnalyzeResponse Existing Embedding and Detect RPCs are reused (face image in PredictOptions.Images / DetectOptions.src) for face embedding and face detection respectively. New HTTP endpoints under /v1/face/: * verify — 1:1 image pair same-person decision * analyze — per-face age + gender (emotion/race reserved) * register — 1:N enrollment; stores embedding in vector store * identify — 1:N recognition; detect → embed → StoresFind * forget — remove a registered face by opaque ID Service layer (core/services/facerecognition/) introduces a `Registry` interface with one in-memory `storeRegistry` impl backed by LocalAI's existing local-store gRPC vector backend. HTTP handlers depend on the interface, not on StoresSet/StoresFind directly, so a persistent PostgreSQL/pgvector implementation can be slotted in via a single constructor change in core/application (TODO marker in the package doc). New usecase flag FLAG_FACE_RECOGNITION; insightface is also wired into FLAG_DETECTION so /v1/detection works for face bounding boxes. Gallery (backend/index.yaml) ships three entries: * insightface-buffalo-l — SCRFD-10GF + ArcFace R50 + genderage (~326MB pre-baked; non-commercial research use only) * insightface-opencv — YuNet + SFace (~40MB pre-baked; Apache 2.0) * insightface-buffalo-s — SCRFD-500MF + MBF (runtime download; non-commercial) Python backend (backend/python/insightface/): * engines.py — FaceEngine protocol with InsightFaceEngine and OnnxDirectEngine; resolves model paths relative to the backend directory so the same gallery config works in docker-scratch and in the e2e-backends rootfs-extraction harness. * backend.py — gRPC servicer implementing Health, LoadModel, Status, Embedding, Detect, FaceVerify, FaceAnalyze. * install.sh — pre-bakes buffalo_l + OpenCV YuNet/SFace inside the backend directory so first-run is offline-clean (the final scratch image only preserves files under /<backend>/). * test.py — parametrized unit tests over both engines. Tests: * Registry unit tests (go test -race ./core/services/facerecognition/...) — in-memory fake grpc.Backend, table-driven, covers register/ identify/forget/error paths + concurrent access. * tests/e2e-backends/backend_test.go extended with face caps (face_detect, face_embed, face_verify, face_analyze); relative ordering + configurable verifyCeiling per engine. * Makefile targets: test-extra-backend-insightface-buffalo-l, -opencv, and the -all aggregate. * CI: .github/workflows/test-extra.yml gains tests-insightface-grpc, auto-triggered by changes under backend/python/insightface/. Docs: * docs/content/features/face-recognition.md — feature page with license table, quickstart (defaults to the commercial-safe model), models matrix, API reference, 1:N workflow, storage caveats. * Cross-refs in object-detection.md, stores.md, embeddings.md, and whats-new.md. * Contributor README at backend/python/insightface/README.md. Verified end-to-end: * buffalo_l: 6/6 specs (health, load, face_detect, face_embed, face_verify, face_analyze). * opencv: 5/5 specs (same minus face_analyze — SFace has no demographic head; correctly skipped via BACKEND_TEST_CAPS). Assisted-by: Claude:claude-opus-4-7 * fix(face-recognition): move engine selection to model gallery, collapse backend entries The previous commit put engine/model_pack options on backend gallery entries (`backend/index.yaml`). That was wrong — `GalleryBackend` (core/gallery/backend_types.go:32) has no `options` field, so the YAML decoder silently dropped those keys and all three "different insightface-*" backend entries resolved to the same container image with no distinguishing configuration. Correct split: * `backend/index.yaml` now has ONE `insightface` backend entry shipping the CPU + CUDA 12 container images. The Python backend bundles both the non-commercial insightface model packs (buffalo_l / buffalo_s) and the commercial-safe OpenCV Zoo weights (YuNet + SFace); the active engine is selected at LoadModel time via `options: ["engine:..."]`. * `gallery/index.yaml` gains three model entries — `insightface-buffalo-l`, `insightface-opencv`, `insightface-buffalo-s` — each setting the appropriate `overrides.backend` + `overrides.options` so installing one actually gives the user the intended engine. This matches how `rfdetr-base` lives in the model gallery against the `rfdetr` backend. The earlier e2e tests passed despite this bug because the Makefile targets pass `BACKEND_TEST_OPTIONS` directly to LoadModel via gRPC, bypassing any gallery resolution entirely. No code changes needed. Assisted-by: Claude:claude-opus-4-7 * feat(face-recognition): cover all supported models in the gallery + drop weight baking Follows up on the model-gallery split: adds entries for every model configuration either engine actually supports, and switches weight delivery from image-baked to LocalAI's standard gallery mechanism. Gallery now has seven `insightface-*` model entries (gallery/index.yaml): insightface (family) — non-commercial research use • buffalo-l (326MB) — SCRFD-10GF + ResNet50 + genderage, default • buffalo-m (313MB) — SCRFD-2.5GF + ResNet50 + genderage • buffalo-s (159MB) — SCRFD-500MF + MBF + genderage • buffalo-sc (16MB) — SCRFD-500MF + MBF, recognition only (no landmarks, no demographics — analyze returns empty attributes) • antelopev2 (407MB) — SCRFD-10GF + ResNet100@Glint360K + genderage OpenCV Zoo family — Apache 2.0 commercial-safe • opencv — YuNet + SFace fp32 (~40MB) • opencv-int8 — YuNet + SFace int8 (~12MB, ~3x smaller, faster on CPU) Model weights are no longer baked into the backend image. The image now ships only the Python runtime + libraries (~275MB content size, ~1.18GB disk vs ~1.21GB when weights were baked). Weights flow through LocalAI's gallery mechanism: * OpenCV variants list `files:` with ONNX URIs + SHA-256, so `local-ai models install insightface-opencv` pulls them into the models directory exactly like any other gallery-managed model. * insightface packs (upstream distributes .zip archives only, not individual ONNX files) auto-download on first LoadModel via FaceAnalysis' built-in machinery, rooted at the LocalAI models directory so they live alongside everything else — same pattern `rfdetr` uses with `inference.get_model()`. Backend changes (backend/python/insightface/): * backend.py — LoadModel propagates `ModelOptions.ModelPath` (the LocalAI models directory) to engines via a `_model_dir` hint. This replaces the earlier ModelFile-dirname approach; ModelPath is the canonical "models directory" variable set by the Go loader (pkg/model/initializers.go:144) and is always populated. * engines.py::_resolve_model_path — picks up `model_dir` and searches it (plus basename-in-model-dir) before falling back to the dev script-dir. This is how OnnxDirectEngine finds gallery-downloaded YuNet/SFace files by filename only. * engines.py::_flatten_insightface_pack — new helper that works around an upstream packaging inconsistency: buffalo_l/s/sc zips expand flat, but buffalo_m and antelopev2 zips wrap their ONNX files in a redundant `<name>/` directory. insightface's own loader looks one level too shallow and fails. We call `ensure_available()` explicitly, flatten if nested, then hand to FaceAnalysis. * engines.py::InsightFaceEngine.prepare — root-resolution order now includes the `_model_dir` hint so packs download into the LocalAI models directory by default. * install.sh — no longer pre-downloads any weights. Everything is gallery-managed now. * smoke.py (new) — parametrized smoke test that iterates over every gallery configuration, simulating the LocalAI install flow (creates a models dir, fetches OpenCV files with checksum verification, lets insightface auto-download its packs), then runs detect + embed + verify (+ analyze where supported) through the in-process BackendServicer. * test.py — OnnxDirectEngineTest no longer hardcodes `/models/opencv/` paths; downloads ONNX files to a temp dir at setUpClass time and passes ModelPath accordingly. Registry change (core/services/facerecognition/store_registry.go): * `dim=0` in NewStoreRegistry now means "accept whatever dimension arrives" — needed because the backend supports 512-d ArcFace/MBF and 128-d SFace via the same Registry. A non-zero dim still fails fast with ErrDimensionMismatch. * core/application plumbs `faceEmbeddingDim = 0`, explaining the rationale in the comment. Backend gallery description updated to reflect that the image carries no weights — it's just Python + engines. Smoke-tested all 7 configurations against the rebuilt image (with the flatten fix applied), exit 0: PASS: insightface-buffalo-l faces=6 dim=512 same-dist=0.000 PASS: insightface-buffalo-sc faces=6 dim=512 same-dist=0.000 PASS: insightface-buffalo-s faces=6 dim=512 same-dist=0.000 PASS: insightface-buffalo-m faces=6 dim=512 same-dist=0.000 PASS: insightface-antelopev2 faces=6 dim=512 same-dist=0.000 PASS: insightface-opencv faces=6 dim=128 same-dist=0.000 PASS: insightface-opencv-int8 faces=6 dim=128 same-dist=0.000 7/7 passed Assisted-by: Claude:claude-opus-4-7 * fix(face-recognition): pre-fetch OpenCV ONNX for e2e target; drop stale pre-baked claim CI regression from the previous commit: I moved OpenCV Zoo weight delivery to LocalAI's gallery `files:` mechanism, but the test-extra-backend-insightface-opencv target was still passing relative paths `detector_onnx:models/opencv/yunet.onnx` in BACKEND_TEST_OPTIONS. The e2e suite drives LoadModel directly over gRPC without going through the gallery, so those relative paths resolved to nothing and OpenCV's ONNXImporter failed: LoadModel failed: Failed to load face engine: OpenCV(4.13.0) ... Can't read ONNX file: models/opencv/yunet.onnx Fix: add an `insightface-opencv-models` prerequisite target that fetches the two ONNX files (YuNet + SFace) to a deterministic host cache at /tmp/localai-insightface-opencv-cache/, verifies SHA-256, and skips the download on re-runs. The opencv test target depends on it and passes absolute paths in BACKEND_TEST_OPTIONS, so the backend finds the files via its normal absolute-path resolution branch. Also refresh the buffalo_l comment: it no longer says "pre-baked" (nothing is — the pack auto-downloads from upstream's GitHub release on first LoadModel, same as in CI). Locally verified: `make test-extra-backend-insightface-opencv` passes 5/5 specs (health, load, face_detect, face_embed, face_verify). Assisted-by: Claude:claude-opus-4-7 * feat(face-recognition): add POST /v1/face/embed + correct /v1/embeddings docs The docs promised that /v1/embeddings returns face vectors when you send an image data-URI. That was never true: /v1/embeddings is OpenAI-compatible and text-only by contract — its handler goes through `core/backend/embeddings.go::ModelEmbedding`, which sets `predictOptions.Embeddings = s` (a string of TEXT to embed) and never populates `predictOptions.Images[]`. The Python backend's Embedding gRPC method does handle Images[] (that's how /v1/face/register reaches it internally via `backend.FaceEmbed`), but the HTTP embeddings endpoint wasn't wired to populate it. Rather than overload /v1/embeddings with image-vs-text detection — messy, and the endpoint is OpenAI-compatible by design — add a dedicated /v1/face/embed endpoint that wraps `backend.FaceEmbed` (already used internally by /v1/face/register and /v1/face/identify). Matches LocalAI's convention of a dedicated path per non-standard flow (/v1/rerank, /v1/detection, /v1/face/verify etc.). Response: { "embedding": [<dim> floats, L2-normed], "dim": int, // 512 for ArcFace R50 / MBF, 128 for SFace "model": "<name>" } Live-tested on the opencv engine: returns a 128-d L2-normalized vector (sum(x^2) = 1.0000). Sentinel in docs updated to note /v1/embeddings is text-only and point image users at /v1/face/embed instead. Assisted-by: Claude:claude-opus-4-7 * fix(http): map malformed image input + gRPC status codes to proper 4xx Image-input failures on LocalAI's single-image endpoints (/v1/detection, /v1/face/{verify,analyze,embed,register,identify}) have historically returned 500 — even when the client was the one who sent garbage. Classic example: you POST an "image" that isn't a URL, isn't a data-URI, and isn't a valid JPEG/PNG — the server shouldn't claim that's its fault. Two helpers land in core/http/endpoints/localai/images.go and every single-image handler is switched over: * decodeImageInput(s) Wraps utils.GetContentURIAsBase64 and turns any failure (invalid URL, not a data-URI, download error, etc.) into echo.NewHTTPError(400, "invalid image input: ..."). * mapBackendError(err) Inspects the gRPC status on a backend call error and maps: INVALID_ARGUMENT → 400 Bad Request NOT_FOUND → 404 Not Found FAILED_PRECONDITION → 412 Precondition Failed Unimplemented → 501 Not Implemented All other codes fall through unchanged (still 500). Before, my 1×1 PNG error-path test returned: HTTP 500 "rpc error: code = InvalidArgument desc = failed to decode one or both images" After: HTTP 400 "failed to decode one or both images" Scope-limited to the LocalAI single-image endpoints. The multi-modal paths (middleware/request.go, openresponses/responses.go, openai/realtime.go) intentionally log-and-skip individual media parts when decoding fails — different design intent (graceful degradation of a multi-part message), not a 400-worthy failure. Left untouched. Live-verified: every error case in /tmp/face_errors.py now returns 4xx with a meaningful message; the "image with no face (1x1 PNG)" case specifically went from 500 → 400. Assisted-by: Claude:claude-opus-4-7 * refactor(face-recognition): insightface packs go through gallery files:, drop FaceAnalysis Follows up on the discovery that LocalAI's gallery `files:` mechanism handles archives (zip, tar.gz, …) via mholt/archiver/v3 — the rhasspy piper voices use exactly this pattern. Insightface packs are zip archives, so we can now deliver them the same way every other gallery-managed model gets delivered: declaratively, checksum-verified, through LocalAI's standard download+extract pipeline. Two changes: 1. Gallery (gallery/index.yaml) — every insightface-* entry gains a `files:` list with the pack zip's URI + SHA-256. `local-ai models install insightface-buffalo-l` now fetches the zip, verifies the hash, and extracts it into the models directory. No more reliance on insightface's library-internal `ensure_available()` auto-download or its hardcoded `BASE_REPO_URL`. 2. InsightFaceEngine (backend/python/insightface/engines.py) — drops the FaceAnalysis wrapper and drives insightface's `model_zoo` directly. The ~50 lines FaceAnalysis provides — glob ONNX files, route each through `model_zoo.get_model()`, build a `{taskname: model}` dict, loop per-face at inference — are reimplemented in `InsightFaceEngine`. The actual inference classes (RetinaFace, ArcFaceONNX, Attribute, Landmark) are still insightface's — we only replicate the glue, so drift risk against upstream is minimal. Why drop FaceAnalysis: it hard-codes a `<root>/models/<name>/*.onnx` layout that doesn't match what LocalAI's zip extraction produces. LocalAI unpacks archives flat into `<models_dir>`. Upstream packs are inconsistent — buffalo_l/s/sc ship ONNX at the zip root (lands at `<models_dir>/*.onnx`), buffalo_m/antelopev2 wrap in a redundant `<name>/` dir (lands at `<models_dir>/<name>/*.onnx`). The new `_locate_insightface_pack` helper searches both locations plus legacy paths and returns whichever has ONNX files. Replaces the earlier `_flatten_insightface_pack` helper (which tried to fight FaceAnalysis's layout expectations; now we just find the files wherever they are). Net effect for users: install once via LocalAI's managed flow, weights live alongside every other model, progress shows in the jobs endpoint, no first-load network call. Same API surface, cleaner plumbing. Assisted-by: Claude:claude-opus-4-7 * fix(face-recognition): CI's insightface e2e path needs the pack pre-fetched The e2e suite drives LoadModel over gRPC without going through LocalAI's gallery flow, so the engine's `_model_dir` option (normally populated from ModelPath) is empty. Previously the insightface target relied on FaceAnalysis auto-download to paper over this, but we dropped FaceAnalysis in favor of direct model_zoo calls — so the buffalo_l target started failing at LoadModel with "no insightface pack found". Mirror the opencv target's pre-fetch pattern: download buffalo_sc.zip (same SHA as the gallery entry), extract it on the host, and pass `root:<dir>` so the engine locates the pack without needing ModelPath. Switched to buffalo_sc (smallest pack, ~16MB) to keep CI fast; it covers the same insightface engine code path as buffalo_l. Face analyze cap dropped since buffalo_sc has no age/gender head. Assisted-by: Claude:claude-opus-4-7[1m] * feat(face-recognition): surface face-recognition in advertised feature maps The six /v1/face/* endpoints were missing from every place LocalAI advertises its feature surface to clients: * api_instructions — the machine-readable capability index at GET /api/instructions. Added `face-recognition` as a dedicated instruction area with an intro that calls out the in-memory registry caveat and the /v1/face/embed vs /v1/embeddings split. * auth/permissions — added FeatureFaceRecognition constant, routed all six face endpoints through it so admins can gate them per-user like any other API feature. Default ON (matches the other API features). * React UI capabilities — CAP_FACE_RECOGNITION symbol mapped to FLAG_FACE_RECOGNITION. Declared only for now; the Face page is a follow-up (noted in the plan). Instruction count bumped 9 → 10; test updated. Assisted-by: Claude:claude-opus-4-7[1m] * docs(agents): capture advertising-surface steps in the endpoint guide Before this change, adding a new /v1/* endpoint reliably missed one or more of: the swagger @Tags annotation, the /api/instructions registry, the auth RouteFeatureRegistry, and the React UI CAP_* symbol. The endpoint would work but be invisible to API consumers, admins, and the UI — and nothing in the existing docs said to look in those places. Extend .agents/api-endpoints-and-auth.md with a new "Advertising surfaces" section covering all four surfaces (swagger tags, /api/ instructions, capabilities.js, docs/), and expand the closing checklist so it's impossible to ship a feature without visiting each one. Hoist a one-liner reminder into AGENTS.md's Quick Reference so agents skim it before diving in. Assisted-by: Claude:claude-opus-4-7[1m]	2026-04-22 21:55:41 +02:00
Ettore Di Giacinto	87e6de1989	feat: wire transcription for llama.cpp, add streaming support (#9353) ... Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-04-14 16:13:40 +02:00
Ettore Di Giacinto	d67623230f	feat(vllm): parity with llama.cpp backend (#9328) ... * fix(schema): serialize ToolCallID and Reasoning in Messages.ToProto The ToProto conversion was dropping tool_call_id and reasoning_content even though both proto and Go fields existed, breaking multi-turn tool calling and reasoning passthrough to backends. * refactor(config): introduce backend hook system and migrate llama-cpp defaults Adds RegisterBackendHook/runBackendHooks so each backend can register default-filling functions that run during ModelConfig.SetDefaults(). Migrates the existing GGUF guessing logic into hooks_llamacpp.go, registered for both 'llama-cpp' and the empty backend (auto-detect). Removes the old guesser.go shim. * feat(config): add vLLM parser defaults hook and importer auto-detection Introduces parser_defaults.json mapping model families to vLLM tool_parser/reasoning_parser names, with longest-pattern-first matching. The vllmDefaults hook auto-fills tool_parser and reasoning_parser options at load time for known families, while the VLLMImporter writes the same values into generated YAML so users can review and edit them. Adds tests covering MatchParserDefaults, hook registration via SetDefaults, and the user-override behavior. * feat(vllm): wire native tool/reasoning parsers + chat deltas + logprobs - Use vLLM's ToolParserManager/ReasoningParserManager to extract structured output (tool calls, reasoning content) instead of reimplementing parsing - Convert proto Messages to dicts and pass tools to apply_chat_template - Emit ChatDelta with content/reasoning_content/tool_calls in Reply - Extract prompt_tokens, completion_tokens, and logprobs from output - Replace boolean GuidedDecoding with proper GuidedDecodingParams from Grammar - Add TokenizeString and Free RPC methods - Fix missing `time` import used by load_video() * feat(vllm): CPU support + shared utils + vllm-omni feature parity - Split vllm install per acceleration: move generic `vllm` out of requirements-after.txt into per-profile after files (cublas12, hipblas, intel) and add CPU wheel URL for cpu-after.txt - requirements-cpu.txt now pulls torch==2.7.0+cpu from PyTorch CPU index - backend/index.yaml: register cpu-vllm / cpu-vllm-development variants - New backend/python/common/vllm_utils.py: shared parse_options, messages_to_dicts, setup_parsers helpers (used by both vllm backends) - vllm-omni: replace hardcoded chat template with tokenizer.apply_chat_template, wire native parsers via shared utils, emit ChatDelta with token counts, add TokenizeString and Free RPCs, detect CPU and set VLLM_TARGET_DEVICE - Add test_cpu_inference.py: standalone script to validate CPU build with a small model (Qwen2.5-0.5B-Instruct) * fix(vllm): CPU build compatibility with vllm 0.14.1 Validated end-to-end on CPU with Qwen2.5-0.5B-Instruct (LoadModel, Predict, TokenizeString, Free all working). - requirements-cpu-after.txt: pin vllm to 0.14.1+cpu (pre-built wheel from GitHub releases) for x86_64 and aarch64. vllm 0.14.1 is the newest CPU wheel whose torch dependency resolves against published PyTorch builds (torch==2.9.1+cpu). Later vllm CPU wheels currently require torch==2.10.0+cpu which is only available on the PyTorch test channel with incompatible torchvision. - requirements-cpu.txt: bump torch to 2.9.1+cpu, add torchvision/torchaudio so uv resolves them consistently from the PyTorch CPU index. - install.sh: add --index-strategy=unsafe-best-match for CPU builds so uv can mix the PyTorch index and PyPI for transitive deps (matches the existing intel profile behaviour). - backend.py LoadModel: vllm >= 0.14 removed AsyncLLMEngine.get_model_config so the old code path errored out with AttributeError on model load. Switch to the new get_tokenizer()/tokenizer accessor with a fallback to building the tokenizer directly from request.Model. * fix(vllm): tool parser constructor compat + e2e tool calling test Concrete vLLM tool parsers override the abstract base's __init__ and drop the tools kwarg (e.g. Hermes2ProToolParser only takes tokenizer). Instantiating with tools= raised TypeError which was silently caught, leaving chat_deltas.tool_calls empty. Retry the constructor without the tools kwarg on TypeError — tools aren't required by these parsers since extract_tool_calls finds tool syntax in the raw model output directly. Validated with Qwen/Qwen2.5-0.5B-Instruct + hermes parser on CPU: the backend correctly returns ToolCallDelta{name='get_weather', arguments='{"location": "Paris, France"}'} in ChatDelta. test_tool_calls.py is a standalone smoke test that spawns the gRPC backend, sends a chat completion with tools, and asserts the response contains a structured tool call. * ci(backend): build cpu-vllm container image Add the cpu-vllm variant to the backend container build matrix so the image registered in backend/index.yaml (cpu-vllm / cpu-vllm-development) is actually produced by CI. Follows the same pattern as the other CPU python backends (cpu-diffusers, cpu-chatterbox, etc.) with build-type='' and no CUDA. backend_pr.yml auto-picks this up via its matrix filter from backend.yml. * test(e2e-backends): add tools capability + HF model name support Extends tests/e2e-backends to cover backends that: - Resolve HuggingFace model ids natively (vllm, vllm-omni) instead of loading a local file: BACKEND_TEST_MODEL_NAME is passed verbatim as ModelOptions.Model with no download/ModelFile. - Parse tool calls into ChatDelta.tool_calls: new "tools" capability sends a Predict with a get_weather function definition and asserts the Reply contains a matching ToolCallDelta. Uses UseTokenizerTemplate with OpenAI-style Messages so the backend can wire tools into the model's chat template. - Need backend-specific Options[]: BACKEND_TEST_OPTIONS lets a test set e.g. "tool_parser:hermes,reasoning_parser:qwen3" at LoadModel time. Adds make target test-extra-backend-vllm that: - docker-build-vllm - loads Qwen/Qwen2.5-0.5B-Instruct - runs health,load,predict,stream,tools with tool_parser:hermes Drops backend/python/vllm/test_{cpu_inference,tool_calls}.py — those standalone scripts were scaffolding used while bringing up the Python backend; the e2e-backends harness now covers the same ground uniformly alongside llama-cpp and ik-llama-cpp. * ci(test-extra): run vllm e2e tests on CPU Adds tests-vllm-grpc to the test-extra workflow, mirroring the llama-cpp and ik-llama-cpp gRPC jobs. Triggers when files under backend/python/vllm/ change (or on run-all), builds the local-ai vllm container image, and runs the tests/e2e-backends harness with BACKEND_TEST_MODEL_NAME=Qwen/Qwen2.5-0.5B-Instruct, tool_parser:hermes, and the tools capability enabled. Uses ubuntu-latest (no GPU) — vllm runs on CPU via the cpu-vllm wheel we pinned in requirements-cpu-after.txt. Frees disk space before the build since the docker image + torch + vllm wheel is sizeable. * fix(vllm): build from source on CI to avoid SIGILL on prebuilt wheel The prebuilt vllm 0.14.1+cpu wheel from GitHub releases is compiled with SIMD instructions (AVX-512 VNNI/BF16 or AMX-BF16) that not every CPU supports. GitHub Actions ubuntu-latest runners SIGILL when vllm spawns the model_executor.models.registry subprocess for introspection, so LoadModel never reaches the actual inference path. - install.sh: when FROM_SOURCE=true on a CPU build, temporarily hide requirements-cpu-after.txt so installRequirements installs the base deps + torch CPU without pulling the prebuilt wheel, then clone vllm and compile it with VLLM_TARGET_DEVICE=cpu. The resulting binaries target the host's actual CPU. - backend/Dockerfile.python: accept a FROM_SOURCE build-arg and expose it as an ENV so install.sh sees it during `make`. - Makefile docker-build-backend: forward FROM_SOURCE as --build-arg when set, so backends that need source builds can opt in. - Makefile test-extra-backend-vllm: call docker-build-vllm via a recursive $(MAKE) invocation so FROM_SOURCE flows through. - .github/workflows/test-extra.yml: set FROM_SOURCE=true on the tests-vllm-grpc job. Slower but reliable — the prebuilt wheel only works on hosts that share the build-time SIMD baseline. Answers 'did you test locally?': yes, end-to-end on my local machine with the prebuilt wheel (CPU supports AVX-512 VNNI). The CI runner CPU gap was not covered locally — this commit plugs that gap. * ci(vllm): use bigger-runner instead of source build The prebuilt vllm 0.14.1+cpu wheel requires SIMD instructions (AVX-512 VNNI/BF16) that stock ubuntu-latest GitHub runners don't support — vllm.model_executor.models.registry SIGILLs on import during LoadModel. Source compilation works but takes 30-40 minutes per CI run, which is too slow for an e2e smoke test. Instead, switch tests-vllm-grpc to the bigger-runner self-hosted label (already used by backend.yml for the llama-cpp CUDA build) — that hardware has the required SIMD baseline and the prebuilt wheel runs cleanly. FROM_SOURCE=true is kept as an opt-in escape hatch: - install.sh still has the CPU source-build path for hosts that need it - backend/Dockerfile.python still declares the ARG + ENV - Makefile docker-build-backend still forwards the build-arg when set Default CI path uses the fast prebuilt wheel; source build can be re-enabled by exporting FROM_SOURCE=true in the environment. * ci(vllm): install make + build deps on bigger-runner bigger-runner is a bare self-hosted runner used by backend.yml for docker image builds — it has docker but not the usual ubuntu-latest toolchain. The make-based test target needs make, build-essential (cgo in 'go test'), and curl/unzip (the Makefile protoc target downloads protoc from github releases). protoc-gen-go and protoc-gen-go-grpc come via 'go install' in the install-go-tools target, which setup-go makes possible. * ci(vllm): install libnuma1 + libgomp1 on bigger-runner The vllm 0.14.1+cpu wheel ships a _C C++ extension that dlopens libnuma.so.1 at import time. When the runner host doesn't have it, the extension silently fails to register its torch ops, so EngineCore crashes on init_device with: AttributeError: '_OpNamespace' '_C_utils' object has no attribute 'init_cpu_threads_env' Also add libgomp1 (OpenMP runtime, used by torch CPU kernels) to be safe on stripped-down runners. * feat(vllm): bundle libnuma/libgomp via package.sh The vllm CPU wheel ships a _C extension that dlopens libnuma.so.1 at import time; torch's CPU kernels in turn use libgomp.so.1 (OpenMP). Without these on the host, vllm._C silently fails to register its torch ops and EngineCore crashes with: AttributeError: '_OpNamespace' '_C_utils' object has no attribute 'init_cpu_threads_env' Rather than asking every user to install libnuma1/libgomp1 on their host (or every LocalAI base image to ship them), bundle them into the backend image itself — same pattern fish-speech and the GPU libs already use. libbackend.sh adds ${EDIR}/lib to LD_LIBRARY_PATH at run time so the bundled copies are picked up automatically. - backend/python/vllm/package.sh (new): copies libnuma.so.1 and libgomp.so.1 from the builder's multilib paths into ${BACKEND}/lib, preserving soname symlinks. Runs during Dockerfile.python's 'Run backend-specific packaging' step (which already invokes package.sh if present). - backend/Dockerfile.python: install libnuma1 + libgomp1 in the builder stage so package.sh has something to copy (the Ubuntu base image otherwise only has libgomp in the gcc dep chain). - test-extra.yml: drop the workaround that installed these libs on the runner host — with the backend image self-contained, the runner no longer needs them, and the test now exercises the packaging path end-to-end the way a production host would. * ci(vllm): disable tests-vllm-grpc job (heterogeneous runners) Both ubuntu-latest and bigger-runner have inconsistent CPU baselines: some instances support the AVX-512 VNNI/BF16 instructions the prebuilt vllm 0.14.1+cpu wheel was compiled with, others SIGILL on import of vllm.model_executor.models.registry. The libnuma packaging fix doesn't help when the wheel itself can't be loaded. FROM_SOURCE=true compiles vllm against the actual host CPU and works everywhere, but takes 30-50 minutes per run — too slow for a smoke test on every PR. Comment out the job for now. The test itself is intact and passes locally; run it via 'make test-extra-backend-vllm' on a host with the required SIMD baseline. Re-enable when: - we have a self-hosted runner label with guaranteed AVX-512 VNNI/BF16, or - vllm publishes a CPU wheel with a wider baseline, or - we set up a docker layer cache that makes FROM_SOURCE acceptable The detect-changes vllm output, the test harness changes (tests/ e2e-backends + tools cap), the make target (test-extra-backend-vllm), the package.sh and the Dockerfile/install.sh plumbing all stay in place.	2026-04-13 11:00:29 +02:00
Ettore Di Giacinto	706cf5d43c	feat(sam.cpp): add sam.cpp detection backend (#9288) ... Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-04-09 21:49:11 +02:00
Ettore Di Giacinto	85be4ff03c	feat(api): add ollama compatibility (#9284) ... Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-04-09 14:15:14 +02:00
Richard Palethorpe	557d0f0f04	feat(api): Allow coding agents to interactively discover how to control and configure LocalAI (#9084) ... Signed-off-by: Richard Palethorpe <io@richiejp.com>	2026-04-04 15:14:35 +02:00
Ettore Di Giacinto	b7e3589875	fix(anthropic): show null index when not present, default to 0 (#9225) ... Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-04-04 15:13:17 +02:00
Ettore Di Giacinto	59108fbe32	feat: add distributed mode (#9124) ... * feat: add distributed mode (experimental) Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix data races, mutexes, transactions Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactorings Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix events and tool stream in agent chat Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * use ginkgo Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(cron): compute correctly time boundaries avoiding re-triggering Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * enhancements, refactorings Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * do not flood of healthy checks Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * do not list obvious backends as text backends Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * tests fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * refactoring and consolidation Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Drop redundant healthcheck Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * enhancements, refactorings Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-03-30 00:47:27 +02:00
walcz-de	00fcf6936c	fix: implement encoding_format=base64 for embeddings endpoint (#9135) ... The OpenAI Node.js SDK v4+ sends encoding_format=base64 by default. LocalAI previously ignored this parameter and always returned a float JSON array, causing a silent data corruption bug in any Node.js client (AnythingLLM Desktop, LangChain.js, LlamaIndex.TS, …): // What the client does when it expects base64 but receives a float array: Buffer.from(floatArray, 'base64') Node.js treats a non-string first argument as a byte array — each float32 value is truncated to a single byte — and Float32Array then reads those bytes as floats, yielding dims/4 values. Vector databases (Qdrant, pgvector, …) then create collections with the wrong dimension, causing all similarity searches to fail silently. e.g. granite-embedding-107m (384 dims) → 96 stored in Qdrant jina-embeddings-v3 (1024 dims) → 256 stored in Qdrant Changes: - core/schema/prediction.go: add EncodingFormat string field to PredictionOptions so the request parameter is parsed and available throughout the request pipeline - core/schema/openai.go: add EmbeddingBase64 string field to Item; add MarshalJSON so the "embedding" JSON key emits either []float32 or a base64 string depending on which field is populated — all other Item consumers (image, video endpoints) are unaffected - core/http/endpoints/openai/embeddings.go: add floatsToBase64() which packs a float32 slice as little-endian bytes and base64-encodes it; add embeddingItem() helper; both InputToken and InputStrings loops now honour encoding_format=base64 Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-03-25 17:38:07 +01:00
Ettore Di Giacinto	031a36c995	feat: inferencing default, automatic tool parsing fallback and wire min_p (#9092) ... * feat: wire min_p Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat: inferencing defaults Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore(refactor): re-use iterative parser Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore: generate automatically inference defaults from unsloth Instead of trying to re-invent the wheel and maintain here the inference defaults, prefer to consume unsloth ones, and contribute there as necessary. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore: apply defaults also to models installed via gallery Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore: be consistent and apply fallback to all endpoint Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-03-22 00:57:15 +01:00
Ettore Di Giacinto	f7e8d9e791	feat(quantization): add quantization backend (#9096) ... Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-03-22 00:56:34 +01:00
Ettore Di Giacinto	d9c1db2b87	feat: add (experimental) fine-tuning support with TRL (#9088) ... * feat: add fine-tuning endpoint Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(experimental): add fine-tuning endpoint and TRL support This changeset defines new GRPC signatues for Fine tuning backends, and add TRL backend as initial fine-tuning engine. This implementation also supports exporting to GGUF and automatically importing it to LocalAI after fine-tuning. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * commit TRL backend, stop by killing process Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * move fine-tune to generic features Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * add evals, reorder menu Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Fix tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-03-21 02:08:02 +01:00
Tv	a6d0e29eba	fix(openresponses): do not omit required field ORItemParam.Arguments (#9074) ... See #9047	2026-03-19 22:04:45 +01:00
Tv	8a0edd0809	Always populate ORItemParam.Summary (#9049) ... * fix(openresponses): do not omit required fields summary and id * fix(openresponses): ensure ORItemParam.Summary is never null Normalize Summary to an empty slice at serialization chokepoints (sendSSEEvent, bufferEvent, buildORResponse) so it always serializes as [] instead of null. Closes #9047	2026-03-18 08:45:46 +01:00
Ettore Di Giacinto	8818452d85	feat(ui): MCP Apps, mcp streaming and client-side support (#8947) ... * Revert "fix: Add timeout-based wait for model deletion completion (#8756)" This reverts commit 9e1b0d0c82. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat: add mcp prompts and resources Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): add client-side MCP Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): allow to authenticate MCP servers Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): add MCP Apps Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore: update AGENTS Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * chore: allow to collapse navbar, save state in storage Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(ui): add MCP button also to home page Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(chat): populate string content Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-03-11 07:30:49 +01:00
Ettore Di Giacinto	a026277ab9	feat(mlx-distributed): add new MLX-distributed backend (#8801) ... * feat(mlx-distributed): add new MLX-distributed backend Add new MLX distributed backend with support for both TCP and RDMA for model sharding. This implementation ties in the discovery implementation already in place, and re-uses the same P2P mechanism for the TCP MLX-distributed inferencing. The Auto-parallel implementation is inspired by Exo's ones (who have been added to acknowledgement for the great work!) Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * expose a CLI to facilitate backend starting Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat: make manual rank0 configurable via model configs Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Add missing features from mlx backend Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Apply suggestion from @mudler Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>	2026-03-09 17:29:32 +01:00
BitToby	96efa4fce0	feat: add WebSocket mode support for the response api (#8676) ... * feat: add WebSocket mode support for the response api Signed-off-by: bittoby <218712309+bittoby@users.noreply.github.com> * test: add e2e tests for WebSocket Responses API Signed-off-by: bittoby <218712309+bittoby@users.noreply.github.com> --------- Signed-off-by: bittoby <218712309+bittoby@users.noreply.github.com>	2026-03-06 10:36:59 +00:00
Ettore Di Giacinto	983db7bedc	feat(ui): add model size estimation (#8684) ... Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-02-28 23:03:47 +01:00
Copilot	3ac7301f31	Add sample_rate support to TTS API via post-processing resampling (#8650) ... * Initial plan * Add TTS sample_rate support via AudioResample post-processing Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com>	2026-02-25 16:36:27 +01:00
Lukas Schaefer	ed0bfb8732	fix: rename json_verbose to verbose_json (#8627) ... Signed-off-by: Lukas Schaefer <lukas@lschaefer.xyz>	2026-02-23 17:57:06 +00:00
Ettore Di Giacinto	53276d28e7	feat(musicgen): add ace-step and UI interface (#8396) ... * feat(musicgen): add ace-step and UI interface Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Correctly handle model dir Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Drop auto-download Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Add to models, fixup UIs icons Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fixups Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Update docs Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * l4t13 is incompatbile Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * avoid pinning version for cuda12 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Drop l4t12 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-02-05 12:04:53 +01:00
Dream	10a1e6c74d	feat(whisperx): add whisperx backend for transcription with speaker diarization (#8299) ... * feat(proto): add speaker field to TranscriptSegment for diarization Add speaker field to the gRPC TranscriptSegment message and map it through the Go schema, enabling backends to return speaker labels. Signed-off-by: eureka928 <meobius123@gmail.com> * feat(whisperx): add whisperx backend for transcription with diarization Add Python gRPC backend using WhisperX for speech-to-text with word-level timestamps, forced alignment, and speaker diarization via pyannote-audio when HF_TOKEN is provided. Signed-off-by: eureka928 <meobius123@gmail.com> * feat(whisperx): register whisperx backend in Makefile Signed-off-by: eureka928 <meobius123@gmail.com> * feat(whisperx): add whisperx meta and image entries to index.yaml Signed-off-by: eureka928 <meobius123@gmail.com> * ci(whisperx): add build matrix entries for CPU, CUDA 12/13, and ROCm Signed-off-by: eureka928 <meobius123@gmail.com> * fix(whisperx): unpin torch versions and use CPU index for cpu requirements Address review feedback: - Use --extra-index-url for CPU torch wheels to reduce size - Remove torch version pins, let uv resolve compatible versions Signed-off-by: eureka928 <meobius123@gmail.com> * fix(whisperx): pin torch ROCm variant to fix CI build failure Signed-off-by: eureka928 <meobius123@gmail.com> * fix(whisperx): pin torch CPU variant to fix uv resolution failure Pin torch==2.8.0+cpu so uv resolves the CPU wheel from the extra index instead of picking torch==2.8.0+cu128 from PyPI, which pulls unresolvable CUDA dependencies. Signed-off-by: eureka928 <meobius123@gmail.com> * fix(whisperx): use unsafe-best-match index strategy to fix uv resolution failure uv's default first-match strategy finds torch on PyPI before checking the extra index, causing it to pick torch==2.8.0+cu128 instead of the CPU variant. This makes whisperx's transitive torch dependency unresolvable. Using unsafe-best-match lets uv consider all indexes. Signed-off-by: eureka928 <meobius123@gmail.com> * fix(whisperx): drop +cpu local version suffix to fix uv resolution failure PEP 440 ==2.8.0 matches 2.8.0+cpu from the extra index, avoiding the issue where uv cannot locate an explicit +cpu local version specifier. This aligns with the pattern used by all other CPU backends. Signed-off-by: eureka928 <meobius123@gmail.com> * fix(backends): drop +rocm local version suffixes from hipblas requirements to fix uv resolution uv cannot resolve PEP 440 local version specifiers (e.g. +rocm6.4, +rocm6.3) in pinned requirements. The --extra-index-url already points to the correct ROCm wheel index and --index-strategy unsafe-best-match (set in libbackend.sh) ensures the ROCm variant is preferred. Applies the same fix as 7f5d72e8 (which resolved this for +cpu) across all 14 hipblas requirements files. Signed-off-by: eureka928 <meobius123@gmail.com> Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> Signed-off-by: eureka928 <meobius123@gmail.com> * revert: scope hipblas suffix fix to whisperx only Reverts changes to non-whisperx hipblas requirements files per maintainer review — other backends are building fine with the +rocm local version suffix. Signed-off-by: eureka928 <meobius123@gmail.com> Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> Signed-off-by: eureka928 <meobius123@gmail.com> --------- Signed-off-by: eureka928 <meobius123@gmail.com> Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>	2026-02-02 16:33:12 +01:00
Andres	b6459ddd57	feat(api): Add transcribe response format request parameter & adjust STT backends (#8318) ... * WIP response format implementation for audio transcriptions (cherry picked from commit e271dd764bbc13846accf3beb8b6522153aa276f) Signed-off-by: Andres Smith <andressmithdev@pm.me> * Rework transcript response_format and add more formats (cherry picked from commit 6a93a8f63e2ee5726bca2980b0c9cf4ef8b7aeb8) Signed-off-by: Andres Smith <andressmithdev@pm.me> * Add test and replace go-openai package with official openai go client (cherry picked from commit f25d1a04e46526429c89db4c739e1e65942ca893) Signed-off-by: Andres Smith <andressmithdev@pm.me> * Fix faster-whisper backend and refactor transcription formatting to also work on CLI Signed-off-by: Andres Smith <andressmithdev@pm.me> (cherry picked from commit 69a93977d5e113eb7172bd85a0f918592d3d2168) Signed-off-by: Andres Smith <andressmithdev@pm.me> --------- Signed-off-by: Andres Smith <andressmithdev@pm.me> Co-authored-by: nanoandrew4 <nanoandrew4@gmail.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>	2026-02-01 17:33:17 +01:00
Ettore Di Giacinto	4077aaf978	chore: re-enable e2e tests, fixups anthropic API tools support (#8296) ... * chore(tests): add mock backend e2e tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Fixup anthropic tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * prepare e2e tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Drop repetitive tests Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Drop specific CI workflow Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fixup anthropic issues, move all e2e tests to use mocked backend Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-01-30 12:41:50 +01:00
Ettore Di Giacinto	68dd9765a0	feat(tts): add support for streaming mode (#8291) ... * feat(tts): add support for streaming mode Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * Send first audio, make sure it's 16 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> --------- Signed-off-by: Ettore Di Giacinto <mudler@localai.io>	2026-01-30 11:58:01 +01:00