mirror of
https://github.com/mudler/LocalAI.git
synced 2026-05-16 20:52:08 -04:00
b4fdb41dcca727789941cccb5c5d9982a44582b3
6339 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
LocalAI [bot]
|
b4fdb41dcc |
fix(distributed): cascade-clean stale node_models rows + filter routing by healthy status (#9754)
* fix(distributed): cascade-clean stale node_models on drain and filter routing by healthy status Stale node_models rows (state="loaded") were surviving past the healthy state of their owning node, causing /embeddings (and other inference paths) to dispatch to a backend whose process was gone or drained. The downstream symptom in a live cluster was pgvector rejecting inserts with "vector cannot have more than 16000 dimensions (SQLSTATE 54000)" because the misbehaving backend silently returned a malformed (oversized) tensor; the Models page showed the model as "running" without an associated node, like a stale entry, even though the node was no longer visible in the Nodes view. Two changes here, plus a third in a follow-up commit: - MarkDraining now cascade-deletes node_models rows for the affected node, mirroring MarkOffline. Drains are explicit operator actions — the box has been intentionally taken out of rotation — so clearing the rows stops the Models UI from misreporting and prevents the routing layer from picking those rows if scheduling logic is ever relaxed. In-flight requests already hold their gRPC client through Route() and finish normally; the only observable effect is a non-fatal IncrementInFlight warning, acceptable for a drain. MarkUnhealthy is deliberately left status-only: it fires from managers_distributed / reconciler on a single nats.ErrNoResponders with no retry, so a transient NATS hiccup must not nuke every loaded model and force a full reload on recovery. - FindAndLockNodeWithModel's inner JOIN now filters on backend_nodes.status = healthy in addition to node_models.state = loaded. The previous version relied on the second node-fetch step to reject non-healthy nodes, but a concurrent reader could still pick the same stale row in the same window. Belt-and-braces. - DistributedConfig.PerModelHealthCheck renamed to DisablePerModelHealthCheck and inverted at the call site so per-model gRPC probing is on by default. The probe (now made consecutive-miss aware in a follow-up commit) independently health- checks each model's gRPC address and removes stale node_models rows when the backend has crashed even though the worker's node-level heartbeat is still arriving. Migration: the field had no CLI flag, env var binding, or YAML key in tree (only the bare struct field), so there is no user-facing migration. Anything constructing DistributedConfig in code needs to drop the assignment (default now does the right thing) or invert it. Assisted-by: Claude:claude-opus-4-7 go-vet go-test golangci-lint Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(distributed): require consecutive misses before per-model probe removes a row The per-model gRPC probe used to remove a node_models row on a single failed health check. With the per-model probe now on by default, that made any 5-second gRPC blip (network jitter, a long-running request hogging the worker's gRPC server thread, brief GC pause) trigger a full reload of the affected model — too eager for production. Require perModelMissThreshold (3) consecutive failed probes before removal. At the default 15s tick a model must be unreachable for ~45s before reap; a single successful probe in between resets the streak. Per-(node, model, replica) state tracked under a mutex on the monitor. If the removal call itself fails, the miss counter is left in place so the next tick retries rather than starting the streak over. Tests: - removes stale model via per-model health check after consecutive failures (replaces the single-shot expectation) - preserves model row when an intermittent failure is followed by a success (covers the reset-on-success path and verifies the counter reset by failing twice more without crossing threshold) - newTestHealthMonitor initializes the misses map so direct-construct test helpers don't nil-map-panic in the probe path Assisted-by: Claude:claude-opus-4-7 go-vet go-test golangci-lint 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> |
||
|
Richard Palethorpe
|
0245b33eab |
feat(realtime): Add Liquid Audio s2s model and assistant mode on talk page (#9801)
* feat(liquid-audio): add LFM2.5-Audio any-to-any backend + realtime_audio usecase
Wires LiquidAI's LFM2.5-Audio-1.5B as a self-contained Realtime API model:
single engine handles VAD, transcription, LLM, and TTS in one bidirectional
stream — drop-in alternative to a VAD+STT+LLM+TTS pipeline.
Backend
- backend/python/liquid-audio/ — new Python gRPC backend wrapping the
`liquid-audio` package. Modes: chat / asr / tts / s2s, voice presets,
Load/Predict/PredictStream/AudioTranscription/TTS/VAD/AudioToAudioStream/
Free and StartFineTune/FineTuneProgress/StopFineTune. Runtime monkey-patch
on `liquid_audio.utils.snapshot_download` so absolute local paths from
LocalAI's gallery resolve without a HF round-trip. soundfile in place of
torchaudio.load/save (torchcodec drags NVIDIA NPP we don't bundle).
- backend/backend.proto + pkg/grpc/{backend,client,server,base,embed,
interface}.go — new AudioToAudioStream RPC mirroring AudioTransformStream
(config/frame/control oneof in; typed event+pcm+meta out).
- core/services/nodes/{health_mock,inflight}_test.go — add stubs for the
new RPC to the test fakes.
Config + capabilities
- core/config/backend_capabilities.go — UsecaseRealtimeAudio, MethodAudio
ToAudioStream, UsecaseInfoMap entry, liquid-audio BackendCapability row.
- core/config/model_config.go — FLAG_REALTIME_AUDIO bitmask, ModalityGroups
membership in both speech-input and audio-output groups so a lone flag
still reads as multimodal, GetAllModelConfigUsecases entry, GuessUsecases
branch.
Realtime endpoint
- core/http/endpoints/openai/realtime.go — extract prepareRealtimeConfig()
so the gate is unit-testable; accept realtime_audio models and self-fill
empty pipeline slots with the model's own name (user-pinned slots win).
- core/http/endpoints/openai/realtime_gate_test.go — six specs covering nil
cfg, empty pipeline, legacy pipeline, self-contained realtime_audio,
user-pinned VAD slot, and partial legacy pipeline.
UI + endpoints
- core/http/routes/ui.go — /api/pipeline-models accepts either a legacy
VAD+STT+LLM+TTS pipeline or a realtime_audio model; surfaces a
self_contained flag so the Talk page can collapse the four cards.
- core/http/routes/ui_api.go — realtime_audio in usecaseFilters.
- core/http/routes/ui_pipeline_models_test.go — covers both code paths.
- core/http/react-ui/src/pages/Talk.jsx — self-contained badge instead of
the four-slot grid; rename Edit Pipeline → Edit Model Config; less
pipeline-specific wording.
- core/http/react-ui/src/pages/Models.jsx + locales/en/models.json — new
realtime_audio filter button + i18n.
- core/http/react-ui/src/utils/capabilities.js — CAP_REALTIME_AUDIO.
- core/http/react-ui/src/pages/FineTune.jsx — voice + validation-dataset
fields, surfaced when backend === liquid-audio, plumbed via
extra_options on submit/export/import.
Gallery + importer
- gallery/liquid-audio.yaml — config template with known_usecases:
[realtime_audio, chat, tts, transcript, vad].
- gallery/index.yaml — four model entries (realtime/chat/asr/tts) keyed by
mode option. Fixed pre-existing `transcribe` typo on the asr entry
(loader silently dropped the unknown string → entry never surfaced as a
transcript model).
- gallery/lfm.yaml — function block for the LFM2 Pythonic tool-call format
`<|tool_call_start|>[name(k="v")]<|tool_call_end|>` matching
common_chat_params_init_lfm2 in vendored llama.cpp.
- core/gallery/importers/{liquid-audio,liquid-audio_test}.go — detector
matches LFM2-Audio HF repos (excludes -gguf mirrors); mode/voice
preferences plumbed through to options.
- core/gallery/importers/importers.go — register LiquidAudioImporter
before LlamaCPPImporter.
- pkg/functions/parse_lfm2_test.go — seven specs for the response/argument
regex pair on the LFM2 pythonic format.
Build matrix
- .github/backend-matrix.yml — seven liquid-audio targets (cuda12, cuda13,
l4t-cuda-13, hipblas, intel, cpu amd64, cpu arm64). Jetpack r36 cuda-12
is skipped (Ubuntu 22.04 / Python 3.10 incompatible with liquid-audio's
3.12 floor).
- backend/index.yaml — anchor + 13 image entries.
- Makefile — .NOTPARALLEL, prepare-test-extra, test-extra,
docker-build-liquid-audio.
Docs
- .agents/plans/liquid-audio-integration.md — phased plan; PR-D (real
any-to-any wiring via AudioToAudioStream), PR-E (mid-audio tool-call
detector), PR-G (GGUF entries once upstream llama.cpp PR #18641 lands)
remain.
- .agents/api-endpoints-and-auth.md — expand the capability-surface
checklist with every place a new FLAG_* needs to be registered.
Assisted-by: claude-code:claude-opus-4-7-1m [Claude Code]
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* feat(realtime): function calling + history cap for any-to-any models
Three pieces, all on the realtime_audio path that just landed:
1. liquid-audio backend (backend/python/liquid-audio/backend.py):
- _build_chat_state grows a `tools_prelude` arg.
- new _render_tools_prelude parses request.Tools (the OpenAI Chat
Completions function array realtime.go already serialises) and
emits an LFM2 `<|tool_list_start|>…<|tool_list_end|>` system turn
ahead of the user history. Mirrors gallery/lfm.yaml's `function:`
template so the model sees the same prompt shape whether served
via llama-cpp or here. Without this the backend silently dropped
tools — function calling was wired end-to-end on the Go side but
the model never saw a tool list.
2. Realtime history cap (core/http/endpoints/openai/realtime.go):
- Session grows MaxHistoryItems int; default picked by new
defaultMaxHistoryItems(cfg) — 6 for realtime_audio models (LFM2.5
1.5B degrades quickly past a handful of turns), 0/unlimited for
legacy pipelines composing larger LLMs.
- triggerResponse runs conv.Items through trimRealtimeItems before
building conversationHistory. Helper walks the cut left if it
would orphan a function_call_output, so tool result + call pairs
stay intact.
- realtime_gate_test.go: specs for defaultMaxHistoryItems and
trimRealtimeItems (zero cap, under cap, over cap, tool-call pair
preservation).
3. Talk page (core/http/react-ui/src/pages/Talk.jsx):
- Reuses the chat page's MCP plumbing — useMCPClient hook,
ClientMCPDropdown component, same auto-connect/disconnect effect
pattern. No bespoke tool registry, no new REST endpoints; tools
come from whichever MCP servers the user toggles on, exactly as
on the chat page.
- sendSessionUpdate now passes session.tools=getToolsForLLM(); the
update re-fires when the active server set changes mid-session.
- New response.function_call_arguments.done handler executes via
the hook's executeTool (which round-trips through the MCP client
SDK), then replies with conversation.item.create
{type:function_call_output} + response.create so the model
completes its turn with the tool output. Mirrors chat's
client-side agentic loop, translated to the realtime wire shape.
UI changes require a LocalAI image rebuild (Dockerfile:308-313 bakes
react-ui/dist into the runtime image). Backend.py changes can be
swapped live in /backends/<id>/backend.py + /backend/shutdown.
Assisted-by: claude-code:claude-opus-4-7-1m [Claude Code]
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* feat(realtime): LocalAI Assistant ("Manage Mode") for the Talk page
Mirrors the chat-page metadata.localai_assistant flow so users can ask the
realtime model what's loaded / installed / configured. Tools are run
server-side via the same in-process MCP holder that powers the chat
modality — no transport switch, no proxy, no new wire protocol.
Wire:
- core/http/endpoints/openai/realtime.go:
- RealtimeSessionOptions{LocalAIAssistant,IsAdmin}; isCurrentUserAdmin
helper mirrors chat.go's requireAssistantAccess (no-op when auth
disabled, else requires auth.RoleAdmin).
- Session grows AssistantExecutor mcpTools.ToolExecutor.
- runRealtimeSession, when opts.LocalAIAssistant is set: gate on admin,
fail closed if DisableLocalAIAssistant or the holder has no tools,
DiscoverTools and inject into session.Tools, prepend
holder.SystemPrompt() to instructions.
- Tool-call dispatch loop: when AssistantExecutor.IsTool(name), run
ExecuteTool inproc, append a FunctionCallOutput to conv.Items, skip
the function_call_arguments client emit (the client can't execute
these — it doesn't know about them). After the loop, if any
assistant tool ran, trigger another response so the model speaks the
result. Mirrors chat's agentic loop, driven server-side rather than
via client round-trip.
- core/http/endpoints/openai/realtime_webrtc.go: RealtimeCallRequest
gains `localai_assistant` (JSON omitempty). Handshake calls
isCurrentUserAdmin and builds RealtimeSessionOptions.
- core/http/react-ui/src/pages/Talk.jsx: admin-only "Manage Mode"
checkbox under the Tools dropdown; passes localai_assistant: true to
realtimeApi.call's body, captured in the connect callback's deps.
Mirroring chat's pattern means the in-process MCP tools surface "just
works" for the Talk page without exposing a Streamable-HTTP MCP endpoint
(which was the alternative). Clients with their own MCP servers can
still use the existing ClientMCPDropdown path in parallel; the realtime
handler distinguishes them by AssistantExecutor.IsTool() at dispatch
time.
Assisted-by: claude-code:claude-opus-4-7-1m [Claude Code]
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* feat(realtime): render Manage Mode tool calls in the Talk transcript
Previously the realtime endpoint only emitted response.output_item.added
for the FunctionCall item, and Talk.jsx's switch ignored the event — so
server-side tool runs were invisible in the UI. The model would speak
the result but the user had no way to see what tool was actually
called.
realtime.go: after executing an assistant tool inproc, emit a second
output_item.added/.done pair for the FunctionCallOutput item. Mirrors
the way the chat page displays tool_call + tool_result blocks.
Talk.jsx: handle both response.output_item.added and .done. Render
FunctionCall (with arguments) and FunctionCallOutput (pretty-printed
JSON when possible) as two transcript entries — `tool_call` with the
wrench icon, `tool_result` with the clipboard icon, both in mono-space
secondary-colour. Resets streamingRef after the result so the next
assistant text delta starts a fresh transcript entry instead of
appending to the previous turn.
Assisted-by: claude-code:claude-opus-4-7-1m [Claude Code]
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* refactor(realtime): bound the Manage Mode tool-loop + preserve assistant tools
Fallout from a review pass on the Manage Mode patches:
- Bound the server-side agentic loop. triggerResponse used to recurse on
executedAssistantTool with no cap — a model that kept calling tools
would blow the goroutine stack. New maxAssistantToolTurns = 10 (mirrors
useChat.js's maxToolTurns). Public triggerResponse is now a thin shim
over triggerResponseAtTurn(toolTurn int); recursion increments the
counter and stops at the cap with an xlog.Warn.
- Preserve Manage Mode tools across client session.update. The handler
used to blindly overwrite session.Tools, so toggling a client MCP
server mid-session silently wiped the in-process admin tools. Session
now caches the original AssistantTools slice at session creation and
the session.update handler merges them back in (client names win on
collision — the client is explicit).
- strconv.ParseBool for the localai_assistant query param instead of
hand-rolled "1" || "true". Mirrors LocalAIAssistantFromMetadata.
- Talk.jsx: render both tool_call and tool_result on
response.output_item.done instead of splitting them across .added and
.done. The server's event pairing (added → done) stays correct; the
UI just doesn't need to inspect both phases of the same item. One
switch case instead of two, no behavioural change.
Out of scope (noted for follow-ups): extract a shared assistant-tools
helper between chat.go and realtime.go (duplication is small enough
that two parallel implementations stay readable for now), and an i18n
key for the Manage Mode helper text (Talk.jsx doesn't use i18n
anywhere else yet).
Assisted-by: claude-code:claude-opus-4-7-1m [Claude Code]
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* ci(test-extra): wire liquid-audio backend smoke test
The backend ships test.py + a `make test` target and is listed in
backend-matrix.yml, so scripts/changed-backends.js already writes a
`liquid-audio=true|false` output when files under backend/python/liquid-audio/
change. The workflow just wasn't reading it.
- Expose the `liquid-audio` output on the detect-changes job
- Add a tests-liquid-audio job that runs `make` + `make test` in
backend/python/liquid-audio, gated on the per-backend detect flag
The smoke covers Health() and LoadModel(mode:finetune); fine-tune mode
short-circuits before any HuggingFace download (backend.py:192), so the
job needs neither weights nor a GPU. The full-inference path remains
gated on LIQUID_AUDIO_MODEL_ID, which CI doesn't set.
The four new Go test files (core/gallery/importers/liquid-audio_test.go,
core/http/endpoints/openai/realtime_gate_test.go,
core/http/routes/ui_pipeline_models_test.go, pkg/functions/parse_lfm2_test.go)
are already picked up by the existing test.yml workflow via `make test` →
`ginkgo -r ./pkg/... ./core/...`; their packages all carry RunSpecs entries.
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Richard Palethorpe <io@richiejp.com>
---------
Signed-off-by: Richard Palethorpe <io@richiejp.com>
|
||
|
Andreas Egli
|
a2940e5d47 |
feat: also parse VRAM budget/usage from vulkaninfo (#9800)
Signed-off-by: Andreas Egli <github@kharan.ch> |
||
|
LocalAI [bot]
|
a645c1f4aa |
chore: ⬆️ Update ggml-org/llama.cpp to a9883db8ee021cf16783016a60996d41820b5195 (#9796)
⬆️ Update ggml-org/llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
957619af53 |
chore: ⬆️ Update ikawrakow/ik_llama.cpp to f9a93c37e2fc021760c3c1aa99cf74c73b7591a7 (#9795)
⬆️ Update ikawrakow/ik_llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com>v4.2.3 |
||
|
LocalAI [bot]
|
ad0ab37230 |
docs: ⬆️ update docs version mudler/LocalAI (#9792)
⬆️ Update docs version mudler/LocalAI Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
0b81e36504 |
chore: ⬆️ Update antirez/ds4 to f8b4ed635d559b3a5b44bf2df6a77e21b3e9178f (#9794)
⬆️ Update antirez/ds4 Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
602866a9d8 |
chore: ⬆️ Update ggml-org/whisper.cpp to 338cce1e58133261753243802a0e7a430118866d (#9793)
⬆️ Update ggml-org/whisper.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
Ettore Di Giacinto
|
8521af145f |
ci(merge): source per-arch digests from ci-cache, not local-ai-backends
Follow-up to PR #9781. v4.2.2 (run 25745181433) showed the keepalive anchor in ci-cache wasn't enough on its own: 19 of 37 multiarch merges still failed with "manifest not found" for the same digests we'd just anchored. Quay's manifest GC is per-repository. The anchor tag in ci-cache protects the manifest copy that lives in ci-cache, but the same digest in local-ai-backends is independently tracked and gets reaped because nothing in local-ai-backends references it (push-by-digest=true leaves it untagged). The merge then asks `local-ai-backends@sha256:<digest>` and quay correctly says "not found" in that repo, even though `ci-cache@sha256:<digest>` is alive and well. Empirical confirmation against a live failed digest from v4.2.2: $ docker buildx imagetools inspect quay.io/go-skynet/ci-cache@sha256:05377fe6... Name: quay.io/go-skynet/ci-cache@sha256:05377fe6... MediaType: application/vnd.docker.distribution.manifest.v2+json $ docker buildx imagetools inspect quay.io/go-skynet/local-ai-backends@sha256:05377fe6... ERROR: ... not found Switch the source of the quay merge step to ci-cache. The blobs the manifest references are already accessible from local-ai-backends (verified via direct registry HEAD: HTTP 200 from both repos — the original push cross-mounted blobs at content-addressable storage time and they outlive the per-repo manifest GC). buildx imagetools create republishes the manifest into local-ai-backends, then writes the user-facing manifest list pointing at it. End state is self-contained: the published manifest list references child manifests by digest only, no embedded reference to ci-cache. Dockerhub merge step is unchanged. Dockerhub's GC isn't aggressive enough to reap untagged manifests at the timescales we operate on (verified: localai/localai-backends@<same digest> still resolves cleanly after >24h). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> |
||
|
LocalAI [bot]
|
bc4cd3dd85 |
feat(llama-cpp): bump to 1ec7ba0c, adapt grpc-server, expose new spec-decoding options (#9765)
* chore(llama.cpp): bump to 1ec7ba0c14f33f17e980daeeda5f35b225d41994
Picks up the upstream `spec : parallel drafting support` change
(ggml-org/llama.cpp#22838) which reshapes the speculative-decoding API
and `server_context_impl`.
Adapt the grpc-server wrapper accordingly:
* `common_params_speculative::type` (single enum) became `types`
(`std::vector<common_speculative_type>`). Update both the
"default to draft when a draft model is set" branch and the
`spec_type`/`speculative_type` option parser. The parser now also
tolerates comma-separated lists, mirroring the upstream
`common_speculative_types_from_names` semantics.
* `common_params_speculative_draft::n_ctx` is gone (draft now shares
the target context size). Keep the `draft_ctx_size` option name for
backward compatibility and ignore the value rather than failing.
* `server_context_impl::model` was renamed to `model_tgt`; update the
two reranker / model-metadata call sites.
Replaces #9763. Builds cleanly under the linux/amd64 cpu-llama-cpp
target locally.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat(llama-cpp): expose new speculative-decoding option keys
Upstream `spec : parallel drafting support` (ggml-org/llama.cpp#22838)
adds the `ngram_mod`, `ngram_map_k`, and `ngram_map_k4v` speculative
families and beefs up the draft-model knobs. The previous bump only
adapted the API; this exposes the new fields through the grpc-server
options dictionary so model configs can drive them.
New `options:` keys (all under `backend: llama-cpp`):
ngram_mod (`ngram_mod` type):
spec_ngram_mod_n_min / spec_ngram_mod_n_max / spec_ngram_mod_n_match
ngram_map_k (`ngram_map_k` type):
spec_ngram_map_k_size_n / spec_ngram_map_k_size_m / spec_ngram_map_k_min_hits
ngram_map_k4v (`ngram_map_k4v` type):
spec_ngram_map_k4v_size_n / spec_ngram_map_k4v_size_m /
spec_ngram_map_k4v_min_hits
ngram lookup caches (`ngram_cache` type):
spec_lookup_cache_static / lookup_cache_static
spec_lookup_cache_dynamic / lookup_cache_dynamic
Draft-model tuning (active when `spec_type` is `draft`):
draft_cache_type_k / spec_draft_cache_type_k
draft_cache_type_v / spec_draft_cache_type_v
draft_threads / spec_draft_threads
draft_threads_batch / spec_draft_threads_batch
draft_cpu_moe / spec_draft_cpu_moe (bool flag)
draft_n_cpu_moe / spec_draft_n_cpu_moe (first N MoE layers on CPU)
draft_override_tensor / spec_draft_override_tensor
(comma-separated <tensor regex>=<buffer type>; re-implements upstream's
static parse_tensor_buffer_overrides since it isn't exported)
`spec_type` already accepted comma-separated lists after the previous
commit, matching upstream's `common_speculative_types_from_names`.
Docs: refresh `docs/content/advanced/model-configuration.md` with
per-family tables and a note about multi-type chaining.
Builds locally with `make docker-build-llama-cpp` (linux/amd64
cpu-llama-cpp AVX variant).
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix(turboquant): bridge new llama.cpp spec API to the legacy fork layout
The previous commits in this series adapted backend/cpp/llama-cpp/grpc-server.cpp
to the post-#22838 (parallel drafting) llama.cpp API. The turboquant build
reuses the same grpc-server.cpp through backend/cpp/turboquant/Makefile,
which copies it into turboquant-<flavor>-build/ and runs patch-grpc-server.sh
on the copy. The fork branched before the API refactor, so it errors out on:
* `ctx_server.impl->model_tgt` (fork still has `model`)
* `params.speculative.{ngram_mod,ngram_map_k,ngram_map_k4v,ngram_cache}.*`
(none of these sub-structs exist in the fork)
* `params.speculative.draft.{cache_type_k/v, cpuparams[, _batch].n_threads,
tensor_buft_overrides}` (fork uses the pre-#22397 flat layout)
* `params.speculative.types` vector / `common_speculative_types_from_names`
(fork has a scalar `type` and only the singular helper)
Approach:
1. backend/cpp/llama-cpp/grpc-server.cpp: introduce a single feature switch
`LOCALAI_LEGACY_LLAMA_CPP_SPEC`. When defined, the two `speculative.type[s]`
discriminations (the "default to draft when a draft model is set" branch
and the `spec_type` / `speculative_type` option parser) fall back to the
singular scalar form, and the entire new-option block (ngram_mod / map_k
/ map_k4v / ngram_cache / draft.{cache_type_*, cpuparams*,
tensor_buft_overrides}) is preprocessed out. The macro is *not* defined
in the source tree — stock llama-cpp builds get the full new API.
2. backend/cpp/turboquant/patch-grpc-server.sh: two new patch steps applied
to the per-flavor build copy at turboquant-<flavor>-build/grpc-server.cpp:
- substitute `ctx_server.impl->model_tgt` -> `ctx_server.impl->model`
- inject `#define LOCALAI_LEGACY_LLAMA_CPP_SPEC 1` before the first
`#include`, so the guarded blocks above drop out for the fork build.
Both patches are idempotent and follow the existing sed/awk pattern in
this script (KV cache types, `get_media_marker`, flat speculative
renames). Stock llama-cpp's `grpc-server.cpp` is never touched.
Drop both legacy patches once the turboquant fork rebases past
ggml-org/llama.cpp#22397 / #22838.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix(turboquant): close draft_ctx_size brace inside legacy guard
The previous turboquant fix wrapped the new option-handler blocks in
`#ifndef LOCALAI_LEGACY_LLAMA_CPP_SPEC ... #endif` but placed the guard
in the middle of an `else if` chain — the `} else if` openings of the
new blocks were responsible for closing the previous block's brace.
With the macro defined the new blocks vanish, draft_ctx_size's `{`
loses its closer, the for-loop's `}` is consumed instead, and the
file ends with a stray opening brace — clang reports it as
`function-definition is not allowed here before '{'` on the next
top-level `int main(...)` and `expected '}' at end of input`.
Move the chain split inside the draft_ctx_size branch:
} else if (... "draft_ctx_size") {
// ...
#ifdef LOCALAI_LEGACY_LLAMA_CPP_SPEC
} // legacy: chain ends here
#else
} else if (... "spec_ngram_mod_n_min") { // modern: chain continues
...
} else if (... "draft_override_tensor") {
...
} // closes last branch
#endif
} // closes for-loop
Brace count is now balanced under both preprocessor branches (verified
with `tr -cd '{' | wc -c` against the patched and unpatched outputs).
Local `make docker-build-turboquant` builds the linux/amd64 cpu-llama-cpp
`turboquant-avx` variant cleanly.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix(ci): forward AMDGPU_TARGETS into Dockerfile.turboquant builder-prebuilt
Dockerfile.turboquant's `builder-prebuilt` stage was missing the
`ARG AMDGPU_TARGETS` / `ENV AMDGPU_TARGETS=${AMDGPU_TARGETS}` pair that
`builder-fromsource` already has (and that `Dockerfile.llama-cpp`
mirrors across both stages). When CI uses the prebuilt base image
(quay.io/go-skynet/ci-cache:base-grpc-*, the common path) the build-arg
passed by the workflow never reaches the env inside the compile stage.
backend/cpp/llama-cpp/Makefile:38 (introduced by #9626) errors out on
hipblas builds when AMDGPU_TARGETS is empty, and the turboquant
Makefile reuses backend/cpp/llama-cpp via a sibling build dir, so the
same check fires from turboquant-fallback under BUILD_TYPE=hipblas:
Makefile:38: *** AMDGPU_TARGETS is empty — set it to a comma-separated
list of gfx targets e.g. gfx1100,gfx1101. Stop.
make: *** [Makefile:66: turboquant-fallback] Error 2
The bug is latent on master because the docker layer cache stays warm
across builds — the compile step rarely re-runs from scratch. The
llama.cpp bump in this PR invalidates the cache, so the missing env var
becomes load-bearing and the hipblas turboquant CI job fails.
Mirror the existing pattern from Dockerfile.llama-cpp.
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>
v4.2.2
|
||
|
LocalAI [bot]
|
86a7f6c9fa |
ci: close GC race + cascade-skip + darwin grpc gaps from v4.2.1 (#9781)
* ci: close the GC race + cascade-skip + darwin grpc gaps from v4.2.1
v4.2.1's backend.yml run (#25701862853) exposed three independent issues
on top of the singletons fix shipped in
|
||
|
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> |
||
|
dependabot[bot]
|
a689100d61 |
chore(deps): bump the npm_and_yarn group across 1 directory with 3 updates (#9728)
Bumps the npm_and_yarn group with 3 updates in the /core/http/react-ui directory: [fast-uri](https://github.com/fastify/fast-uri), [hono](https://github.com/honojs/hono) and [ip-address](https://github.com/beaugunderson/ip-address). Updates `fast-uri` from 3.1.0 to 3.1.2 - [Release notes](https://github.com/fastify/fast-uri/releases) - [Commits](https://github.com/fastify/fast-uri/compare/v3.1.0...v3.1.2) Updates `hono` from 4.12.14 to 4.12.18 - [Release notes](https://github.com/honojs/hono/releases) - [Commits](https://github.com/honojs/hono/compare/v4.12.14...v4.12.18) Updates `ip-address` from 10.1.0 to 10.2.0 - [Commits](https://github.com/beaugunderson/ip-address/commits) --- updated-dependencies: - dependency-name: fast-uri dependency-version: 3.1.2 dependency-type: indirect dependency-group: npm_and_yarn - dependency-name: hono dependency-version: 4.12.18 dependency-type: indirect dependency-group: npm_and_yarn - dependency-name: ip-address dependency-version: 10.2.0 dependency-type: indirect dependency-group: npm_and_yarn ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
Andreas Egli
|
03815e3b59 |
fix: parse vulkan VRAM from text (#9669)
* fix: parse vulkan VRAM from text Assisted-by: opencode:gpt-5.5 Signed-off-by: Andreas Egli <github@kharan.ch> * fix: replace string.split with streaming iteration Assisted-by: Opencode:Gemma4 Signed-off-by: Andreas Egli <github@kharan.ch> --------- Signed-off-by: Andreas Egli <github@kharan.ch> |
||
|
dependabot[bot]
|
37991c8a18 |
chore(deps): bump github.com/mudler/edgevpn from 0.31.1 to 0.32.2 (#9773)
Bumps [github.com/mudler/edgevpn](https://github.com/mudler/edgevpn) from 0.31.1 to 0.32.2. - [Release notes](https://github.com/mudler/edgevpn/releases) - [Commits](https://github.com/mudler/edgevpn/compare/v0.31.1...v0.32.2) --- updated-dependencies: - dependency-name: github.com/mudler/edgevpn dependency-version: 0.32.2 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
61c9b187fa |
chore(deps): update charset-normalizer requirement from >=3.4.0 to >=3.4.7 in /backend/python/vllm (#9779)
chore(deps): update charset-normalizer requirement Updates the requirements on [charset-normalizer](https://github.com/jawah/charset_normalizer) to permit the latest version. - [Release notes](https://github.com/jawah/charset_normalizer/releases) - [Changelog](https://github.com/jawah/charset_normalizer/blob/master/CHANGELOG.md) - [Commits](https://github.com/jawah/charset_normalizer/compare/3.4.0...3.4.7) --- updated-dependencies: - dependency-name: charset-normalizer dependency-version: 3.4.7 dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
c66014312e |
chore(deps): bump github.com/fsnotify/fsnotify from 1.9.0 to 1.10.1 (#9778)
Bumps [github.com/fsnotify/fsnotify](https://github.com/fsnotify/fsnotify) from 1.9.0 to 1.10.1. - [Release notes](https://github.com/fsnotify/fsnotify/releases) - [Changelog](https://github.com/fsnotify/fsnotify/blob/main/CHANGELOG.md) - [Commits](https://github.com/fsnotify/fsnotify/compare/v1.9.0...v1.10.1) --- updated-dependencies: - dependency-name: github.com/fsnotify/fsnotify dependency-version: 1.10.1 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
abc2a51641 |
chore(deps): update transformers requirement from >=5.0.0 to >=5.8.0 in /backend/python/transformers (#9775)
chore(deps): update transformers requirement Updates the requirements on [transformers](https://github.com/huggingface/transformers) to permit the latest version. - [Release notes](https://github.com/huggingface/transformers/releases) - [Commits](https://github.com/huggingface/transformers/compare/v5.0.0...v5.8.0) --- updated-dependencies: - dependency-name: transformers dependency-version: 5.8.0 dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
cd7d163178 |
chore(deps): bump github.com/onsi/gomega from 1.39.1 to 1.40.0 (#9774)
Bumps [github.com/onsi/gomega](https://github.com/onsi/gomega) from 1.39.1 to 1.40.0. - [Release notes](https://github.com/onsi/gomega/releases) - [Changelog](https://github.com/onsi/gomega/blob/master/CHANGELOG.md) - [Commits](https://github.com/onsi/gomega/compare/v1.39.1...v1.40.0) --- updated-dependencies: - dependency-name: github.com/onsi/gomega dependency-version: 1.40.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
7aac599deb |
chore(deps): bump github.com/anthropics/anthropic-sdk-go from 1.27.0 to 1.42.0 (#9772)
chore(deps): bump github.com/anthropics/anthropic-sdk-go Bumps [github.com/anthropics/anthropic-sdk-go](https://github.com/anthropics/anthropic-sdk-go) from 1.27.0 to 1.42.0. - [Release notes](https://github.com/anthropics/anthropic-sdk-go/releases) - [Changelog](https://github.com/anthropics/anthropic-sdk-go/blob/main/CHANGELOG.md) - [Commits](https://github.com/anthropics/anthropic-sdk-go/compare/v1.27.0...v1.42.0) --- updated-dependencies: - dependency-name: github.com/anthropics/anthropic-sdk-go dependency-version: 1.42.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
d75173dd2a |
chore(deps): bump actions/download-artifact from 4 to 8 (#9771)
Bumps [actions/download-artifact](https://github.com/actions/download-artifact) from 4 to 8. - [Release notes](https://github.com/actions/download-artifact/releases) - [Commits](https://github.com/actions/download-artifact/compare/v4...v8) --- updated-dependencies: - dependency-name: actions/download-artifact dependency-version: '8' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
9be5310394 |
chore(deps): bump actions/upload-artifact from 4 to 7 (#9770)
Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4 to 7. - [Release notes](https://github.com/actions/upload-artifact/releases) - [Commits](https://github.com/actions/upload-artifact/compare/v4...v7) --- updated-dependencies: - dependency-name: actions/upload-artifact dependency-version: '7' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
dependabot[bot]
|
cdf50fd723 |
chore(deps): bump node from 25-slim to 26-slim (#9769)
Bumps node from 25-slim to 26-slim. --- updated-dependencies: - dependency-name: node dependency-version: 26-slim dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
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>
v4.2.1
|
||
|
LocalAI [bot]
|
78722caedc |
chore: ⬆️ Update ikawrakow/ik_llama.cpp to eb570eb96689c235933b813693ca28ab9d3d26de (#9764)
⬆️ Update ikawrakow/ik_llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
621c612b2d |
ci(bump-deps): register ds4 + move version pin into the Makefile (#9761)
* ci(bump-deps): register ds4 + move version pin into the Makefile The initial ds4 PR (#9758) put the upstream commit pin in backend/cpp/ds4/prepare.sh as a shell variable. The auto-bump bot at .github/bump_deps.sh greps for ^$VAR?= in a Makefile, so DS4_VERSION was invisible to it - other backends (llama-cpp, ik-llama-cpp, turboquant, voxtral, etc.) all pin in their Makefile. This change: - Moves DS4_VERSION?= and DS4_REPO?= to the top of backend/cpp/ds4/Makefile. - Inlines the git init/fetch/checkout recipe into the 'ds4:' target (matches llama-cpp's 'llama.cpp:' target pattern). Directory acts as the target so make only re-clones when missing. - Deletes the now-redundant prepare.sh. - Adds antirez/ds4 + DS4_VERSION + main + backend/cpp/ds4/Makefile to the .github/workflows/bump_deps.yaml matrix so the daily bot opens PRs against this pin. - Updates .agents/ds4-backend.md to point at the Makefile. Verified: $ grep -m1 '^DS4_VERSION?=' backend/cpp/ds4/Makefile DS4_VERSION?=ae302c2fa18cc6d9aefc021d0f27ae03c9ad2fc0 $ make -C backend/cpp/ds4 ds4 # clones into ds4/ at the pin $ make -C backend/cpp/ds4 ds4 # no-op on second invocation make: 'ds4' is up to date. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci: route backend/cpp/ds4/ changes through changed-backends.js scripts/changed-backends.js:inferBackendPath has an explicit branch per cpp dockerfile suffix (ik-llama-cpp, turboquant, llama-cpp). Without a matching branch the function returns null, the backend never lands in the path map, and PR change-detection cannot map "backend/cpp/ds4/X changed" -> "rebuild ds4 image". This is why PR #9761 produced zero ds4 jobs even though it directly edits backend/cpp/ds4/Makefile. Adds the missing branch (Dockerfile.ds4 -> backend/cpp/ds4/), placed before the llama-cpp branch (since both share the .cpp ancestry but ds4 is more specific - same ordering rule documented in .agents/adding-backends.md). Verified with a local Node simulation of the script against this PR's diff: the path map now contains 'ds4 -> backend/cpp/ds4/' and a 'backend/cpp/ds4/Makefile' change correctly triggers the ds4 backend in the rebuild set. Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * docs(adding-backends): harden the two gotchas that bit ds4 Both omissions are silent at the time you ADD a backend - the failure mode only appears later (the bump bot stays silent forever, or the path filter shows up on the next PR that touches your backend with zero CI jobs and looks broken for unrelated reasons). Expanding the `scripts/changed-backends.js` paragraph from a one-liner to a fully worked example, and adding a new sibling paragraph for the `bump_deps.yaml` + Makefile-pin contract. Both call out the specific mistakes from the ds4 timeline (#9758 → #9761) so future contributors can pattern-match on the cause. 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> |
||
|
LocalAI [bot]
|
e3f9de1026 |
docs: ⬆️ update docs version mudler/LocalAI (#9762)
⬆️ Update docs version mudler/LocalAI Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
d892e4af80 |
feat: add ds4 backend (DeepSeek V4 Flash) with tool calls, thinking, KV cache (#9758)
* test(e2e-backends): allow BACKEND_BINARY for native-built backends
Adds an escape hatch for hardware-gated backends (e.g. ds4) where the
model is too large for Docker build context. When BACKEND_BINARY points
at a run.sh produced by 'make -C backend/cpp/<name> package', the suite
skips docker image extraction and drives the binary directly.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* test(e2e-backends): validate BACKEND_BINARY basename + log actual source
Two follow-ups from the
|
||
|
dependabot[bot]
|
5d0f732b16 |
chore(deps): bump the go_modules group across 1 directory with 2 updates (#9759)
Bumps the go_modules group with 2 updates in the / directory: [github.com/gofiber/utils](https://github.com/gofiber/utils) and [github.com/go-git/go-git/v5](https://github.com/go-git/go-git). Updates `github.com/gofiber/utils` from 1.1.0 to 1.2.0 - [Release notes](https://github.com/gofiber/utils/releases) - [Commits](https://github.com/gofiber/utils/compare/v1.1.0...v1.2.0) Updates `github.com/go-git/go-git/v5` from 5.18.0 to 5.19.0 - [Release notes](https://github.com/go-git/go-git/releases) - [Changelog](https://github.com/go-git/go-git/blob/main/HISTORY.md) - [Commits](https://github.com/go-git/go-git/compare/v5.18.0...v5.19.0) --- updated-dependencies: - dependency-name: github.com/gofiber/utils dependency-version: 1.2.0 dependency-type: indirect dependency-group: go_modules - dependency-name: github.com/go-git/go-git/v5 dependency-version: 5.19.0 dependency-type: indirect dependency-group: go_modules ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> |
||
|
Ettore Di Giacinto
|
ea00199554 |
ci: tag every backend digest, including singletons
backend_build.yml pushes by canonical digest only (push-by-digest=true, no tags applied at build time). User-facing tagging happens in backend_merge.yml's `imagetools create` step. Before this commit, scripts/changed-backends.js emitted a merge entry only for tag-suffixes with 2+ legs, so every single-arch backend (CUDA/ROCm/Intel Python images, vLLM, sglang, transformers, diffusers, ...) pushed its digest untagged and stayed that way until quay's GC reaped it. Symptom: tag releases shipped multi-arch backends tagged correctly, but no v<X>-gpu-nvidia-cuda-12-vllm (or any singleton variant) ever appeared in the registry. Changes: - scripts/changed-backends.js drops the `group.length < 2` skip and emits two merge matrices, one per arch class, so each downstream merge job can `needs:` only its corresponding build matrix. - backend.yml splits backend-merge-jobs into multiarch and singlearch variants. The split preserves PR #9746's fix: slow singlearch CUDA builds (~6h) must not gate multiarch merges, or quay's GC reaps the multiarch per-arch digests before they're tagged. - backend_pr.yml mirrors the split. - backend_build.yml renames the digest artifact from `digests<suffix>-<platform-tag>` to `digests<suffix>--<platform-tag-or-"single">`. The `--` separator prevents the merge-side glob from over-matching sibling backends whose tag-suffix is a prefix of ours (e.g. -cpu-vllm vs -cpu-vllm-omni, -cpu-mlx vs -cpu-mlx-audio); the `single` placeholder keeps the name well-formed when platform-tag is empty. - backend_merge.yml updates the download pattern to match. Verified locally: a tag-push event now expands to 36 multiarch merge entries (= 72 builds / 2 legs) and 199 singlearch merge entries (one per singleton, including -gpu-nvidia-cuda-12-vllm at index 24). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> |
||
|
LocalAI [bot]
|
b9e81dbfd4 |
chore: ⬆️ Update ggml-org/llama.cpp to 389ff61d77b5c71cec0cf92fe4e5d01ace80b797 (#9752)
⬆️ Update ggml-org/llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>v4.2.0 |
||
|
Ettore Di Giacinto
|
059c493641 |
ci(darwin): brew reinstall ccache to handle transitive dep drift
Symptom (PR #9752, run 25638825961, job 75256261163):
dyld[11144]: Library not loaded: /opt/homebrew/opt/fmt/lib/libfmt.12.dylib
Referenced from: /opt/homebrew/Cellar/ccache/4.13.5/bin/ccache
Abort trap: 6
Previous fix (commit
|
||
|
LocalAI [bot]
|
19d59102d5 |
feat(whisper-cpp): implement streaming transcription (#9751)
* test(whisper): wire e2e streaming transcription target Adds test-extra-backend-whisper-transcription, mirroring the existing llama-cpp / sherpa-onnx / vibevoice-cpp targets. The generic AudioTranscriptionStream spec at tests/e2e-backends/backend_test.go:644 fails today because backend/go/whisper has no streaming impl - this target is the failing TDD gate that the next phase makes pass. Confirmed RED locally: 3 Passed (health, load, offline transcription), 1 Failed (streaming spec hits its 300s context deadline because the base implementation returns 'unimplemented' but doesn't close the result channel, leaving the gRPC stream open until the client times out). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(whisper-cpp): expose new_segment_callback to the Go side Adds set_new_segment_callback() and a C-side trampoline that whisper.cpp invokes once per new text segment during whisper_full(). The trampoline dispatches (idx_first, n_new, user_data) to a Go function pointer registered via purego.NewCallback - text and timings are pulled by Go through the existing get_segment_text/get_segment_t0/get_segment_t1 getters. Wires the hook only when streaming is actually requested, to avoid a per-segment function-pointer dispatch on the offline path. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * feat(whisper-cpp): implement AudioTranscriptionStream Wires whisper.cpp's new_segment_callback through purego back to Go so the streaming transcription RPC produces real, time-correlated deltas while whisper_full() is still decoding. Each segment becomes one TranscriptStreamResponse{Delta}; whisper_full's return is the TranscriptStreamResponse{FinalResult} carrying the full segment list, language, and duration. Per-call state is tracked in a sync.Map keyed by an atomic counter; the Go callback registered via purego.NewCallback is a singleton, dispatched through user_data. SingleThread today means only one entry is ever live, but the map shape matches the sherpa-onnx TTS callback pattern. The streaming path's final.Text is the literal concat of every emitted delta (a strings.Builder accumulated by onNewSegment) so the e2e invariant `final.Text == concat(deltas)` holds exactly. The first delta has no leading space; subsequent deltas are space-prefixed. The offline AudioTranscription path is unchanged. Closes the gap with sherpa-onnx, vibevoice-cpp, llama-cpp, and tinygrad, which already implement AudioTranscriptionStream. Verified GREEN locally: make test-extra-backend-whisper-transcription passes 4/4 specs (3 Passed initially under RED, +1 streaming spec now). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * test(whisper-cpp): assert progressive multi-segment streaming Drives AudioTranscriptionStream against a real long-audio fixture and asserts len(deltas) >= 2. The generic e2e spec at tests/e2e-backends/backend_test.go:644 only checks len(deltas) >= 1 which is satisfied by both real and faked streaming - this spec is the guardrail that a future "fake" impl can't sneak past. Skipped by default (env-gated, like the cancellation spec); set WHISPER_LIBRARY, WHISPER_MODEL_PATH, and WHISPER_AUDIO_PATH to a 30+ second clip to run. Verified locally with a 55s 5x-JFK concat against ggml-base.en.bin: 1 Passed in 7.3s, deltas >= 2, finalSegmentCount >= 2, concat(deltas) == final.Text. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci(whisper-cpp): add transcription gRPC e2e job Mirrors tests-sherpa-onnx-grpc-transcription / tests-llama-cpp-grpc-transcription. Runs make test-extra-backend-whisper-transcription whenever the whisper backend or the run-all switch fires, so a pin-bump or refactor that breaks streaming transcription gets caught before merge. The whisper output on detect-changes is already emitted by scripts/changed-backends.js (it iterates allBackendPaths); this PR just exposes it as a workflow output and consumes it. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * fix(whisper-cpp): silence errcheck on AudioTranscriptionStream defers golangci-lint runs with new-from-merge-base=origin/master, so the identical defer patterns in the existing offline AudioTranscription path are grandfathered while the new ones in AudioTranscriptionStream trip errcheck. Wrap both defers in `func() { _ = ... }()` to match what errcheck wants without altering behavior. The errors from os.RemoveAll and *os.File.Close are not actionable inside a defer here (we're already returning), matching the offline path's contract. Assisted-by: Claude:claude-opus-4-7 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> |
||
|
LocalAI [bot]
|
4715a68660 |
chore: ⬆️ Update vllm-project/vllm cu130 wheel to 0.20.2 (#9750)
⬆️ Update vllm-project/vllm cu130 wheel Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
28f33be48f |
chore: ⬆️ Update ggml-org/whisper.cpp to c33c5618b72bb345df029b730b36bc0e369845a3 (#9749)
⬆️ Update ggml-org/whisper.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
a435f7cc69 |
chore: ⬆️ Update ikawrakow/ik_llama.cpp to 23127139cb6fa314899c3b5f4935b88b3374c56c (#9748)
⬆️ Update ikawrakow/ik_llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
f6c9c20911 |
chore: ⬆️ Update ggml-org/llama.cpp to 2b2babd1243c67ca811c0a5852cedf92b1a20024 (#9747)
⬆️ Update ggml-org/llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com> |
||
|
Ettore Di Giacinto
|
3f6e493439 |
ci(darwin): install ccache's runtime dylib deps (blake3, hiredis, xxhash, zstd)
Symptom (run 25634195866, job 75244019809): the Configure ccache step
on the Darwin llama-cpp build aborted with:
dyld[5647]: Library not loaded: /opt/homebrew/opt/blake3/lib/libblake3.0.dylib
Referenced from: /opt/homebrew/Cellar/ccache/4.13.5/bin/ccache
Abort trap: 6
The previous Darwin fix (
|
||
|
LocalAI [bot]
|
35f6db8c76 |
ci: split backend-jobs into single-arch and multi-arch matrices (#9746)
Symptom (run 25612992409): backend-merge-jobs failed with
"quay.io/go-skynet/local-ai-backends@sha256:fdbd93ca...: not found"
even though the per-arch build for -cpu-llama-cpp pushed that exact
digest 14h31m earlier.
Root cause: backend-merge-jobs was gated on the WHOLE backend-jobs
matrix (`needs: backend-jobs`). The multi-arch -cpu-llama-cpp legs
finished within 30 min, but a single-arch CUDA-12-llama-cpp slot in
the same matrix queued for ~8h (max-parallel: 8 throttle) and then
took ~6h to build cold. By the time it freed the merge to run, quay's
GC had reaped the per-arch digests pushed by the fast multi-arch legs
the day before.
Fix: split the linux backend matrix in two.
backend-jobs-multiarch - entries with `platform-tag` set (paired
per-arch legs that feed backend-merge-jobs).
backend-jobs-singlearch - entries without `platform-tag` (heavy
standalone builds: CUDA, ROCm, Intel oneAPI, vLLM, sglang, etc.).
backend-merge-jobs now `needs:` only backend-jobs-multiarch. The
multi-arch matrix completes in ~2-3h, well inside quay's GC window.
Heavy single-arch entries keep running independently with no merge
dependency.
scripts/changed-backends.js gains a splitByArch() helper that
partitions filtered entries by whether `platform-tag` is set, and
emits matrix-singlearch + matrix-multiarch + has-backends-singlearch
+ has-backends-multiarch outputs (replacing the previous combined
matrix / has-backends pair). Applied in both the full-matrix and
filtered-matrix code paths. Smoke test: 199 single-arch + 72 multi-
arch + 35 darwin = 271 total entries; 36 merge-matrix entries
(one per multi-arch backend pair). Matches expectation.
Local `make backends/<name>` is unaffected — the script's outputs
only feed CI workflow matrices.
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
|
||
|
Ettore Di Giacinto
|
6113e5a4d0 |
docs(ci-caching): list all paths that retrigger base-images.yml
Now that base-images.yml's master-push trigger includes the install
script and apt-mirror script (commit
|
||
|
Ettore Di Giacinto
|
7fff858408 |
ci(base-images): also trigger rebuild on .docker/install-base-deps.sh changes
base-images.yml's master-push trigger had a path filter listing only backend/Dockerfile.base-grpc-builder and .github/workflows/base-images.yml. That misses .docker/install-base-deps.sh — which is the actual source of truth for what goes into each base image (apt deps, gRPC, conditional CUDA/ROCm/Vulkan installs). The script is bind-mounted into the base Dockerfile at build time; changes to it would change the produced images, but without this path filter, the workflow wouldn't auto-rebuild on those changes. Stale bases would persist until Saturday's cron or a manual workflow_dispatch. Same applies to .docker/apt-mirror.sh, also bind-mounted by the base Dockerfile. Add both to the trigger paths so consumer-affecting changes to either file rebuild the bases automatically. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> |
||
|
LocalAI [bot]
|
6fd21d5cf3 |
docs(agents): update CI caching docs after the GHA-free-tier migration (#9742)
The migration shipped over a sequence of PRs (#9726 → #9727 → #9730 → #9731 → #9737 → #9738 plus a handful of direct-to-master fixes) and left the .agents/ docs significantly out of date. Updated: - .agents/ci-caching.md (significant rewrite) - Cache key shape: now includes per-arch suffix (cache<suffix>-<arch>). - New "Workflow surfaces" overview table. - New "Pre-built base images (base-grpc-*)" section covering the 10 quay.io/go-skynet/ci-cache:base-grpc-* tags, the multi-target Dockerfile pattern (builder-fromsource / builder-prebuilt / aliasing FROM), the BUILDER_BASE_IMAGE → BUILDER_TARGET derivation, the bootstrap-on-branch order for new variants. - New "Per-arch native builds + manifest merge" section: split matrix entries, push-by-digest, backend_merge.yml, why provenance: false matters. - New "Path filter on master push" section: changed-backends.js handles push events via the Compare API; weekly Sunday cron is the safety net for unpinned Python deps. - New "ccache for C++ backend builds" section. - New "Composite actions" section: free-disk-space and setup-build-disk. - New "Concurrency" section documenting the per-PR-per-commit group fix. - Darwin section gains the brew link --overwrite note (after- cache-restore symlinks weren't restored) and the llama-cpp-darwin consolidation context. - "Self-hosted runners" section confirming the matrix is free of arc-runner-set / bigger-runner references except the residual test-extra.yml vibevoice case. - "Touching the cache pipeline" rule list extended (provenance, install-base-deps.sh single-source-of-truth, base-images bootstrap order). - .agents/adding-backends.md - Section 2 title: backend.yml -> backend-matrix.yml (path moved). - New paragraph on per-arch entries (platform-tag + paired matrix rows + auto-firing merge job). - New paragraph on builder-base-image for llama-cpp / ik-llama-cpp / turboquant. - Final checklist line updated accordingly. - .agents/building-and-testing.md - Reference: backend.yml -> backend-matrix.yml. - Note about builder-base-image and BUILDER_TARGET defaulting to builder-fromsource for local builds. - AGENTS.md - One-line description update for ci-caching.md to mention the new infrastructure (per-arch keys, base-grpc-*, manifest-merge, setup-build-disk, path filter). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> Co-authored-by: Ettore Di Giacinto <mudler@localai.io> |
||
|
LocalAI [bot]
|
6cbf69dc29 |
chore: ⬆️ Update ggml-org/llama.cpp to 1e5ad35d560b90a8ac447d149c8f8447ae1fcaa0 (#9739)
⬆️ Update ggml-org/llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com> |
||
|
LocalAI [bot]
|
593f3a8648 |
ci: refactor llama-cpp variant Dockerfiles to consume prebuilt base-grpc images (PR 2/2) (#9738)
* ci(backend_build): plumb builder-base-image and BUILDER_TARGET build-args Adds an optional builder-base-image input. When set, BUILDER_BASE_IMAGE is forwarded as a build-arg AND BUILDER_TARGET=builder-prebuilt is set to select the variant Dockerfile's prebuilt-base stage. When empty, BUILDER_TARGET=builder-fromsource (the default) keeps the existing from-source build path. This makes the prebuilt-base optimization opt-in per matrix entry without breaking local `make backends/<name>` invocations or backends whose Dockerfile doesn't have a prebuilt path. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci(llama-cpp,ik-llama-cpp,turboquant): multi-target Dockerfiles for prebuilt + from-source Restructure the three llama.cpp-derived Dockerfiles so each supports two builder paths in a single file, selected via the BUILDER_TARGET build-arg: BUILDER_TARGET=builder-fromsource (default) - Standalone build: gRPC stage + apt installs + (conditionally) CUDA/ROCm/Vulkan + compile. - Used by `make backends/llama-cpp` locally and any caller that doesn't supply a prebuilt base. BUILDER_TARGET=builder-prebuilt - FROM \${BUILDER_BASE_IMAGE} (one of quay.io/go-skynet/ci-cache: base-grpc-* shipped in PR #9737). - Skips ~25-35 min of gRPC compile + ~5-10 min of toolchain installs. - Used by CI when the matrix entry sets builder-base-image. Final FROM scratch resolves BUILDER_TARGET via an aliasing FROM stage (BuildKit doesn't support variable expansion directly in COPY --from), then COPY --from=builder pulls package output from the chosen path. BuildKit prunes the unreferenced builder, so each build only does the work for the chosen path. The compile RUN is identical between both builder stages, so it's factored into .docker/<name>-compile.sh and bind-mounted into both. ccache mount + cache-id stay per-arch / per-build-type. Local DX preserved: `make backends/llama-cpp` (no extra args) defaults to BUILDER_TARGET=builder-fromsource and works exactly as before. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci(backend.yml,backend_pr.yml): forward builder-base-image from matrix Plumbs the new optional builder-base-image input from matrix into backend_build.yml. backend_build.yml derives BUILDER_TARGET from whether builder-base-image is set, so matrix entries that map to a prebuilt base get the prebuilt path; entries that don't (python/go/ rust backends) fall through to the default builder-fromsource (which their own Dockerfiles don't reference, so it's a no-op for them). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci(backend-matrix): wire builder-base-image to llama-cpp variants For every entry whose Dockerfile is llama-cpp/ik-llama-cpp/turboquant, add a builder-base-image field pointing at the appropriate prebuilt quay.io/go-skynet/ci-cache:base-grpc-* tag. backend_build.yml derives BUILDER_TARGET from this field's presence: non-empty -> builder-prebuilt; empty -> builder-fromsource. So this commit alone activates the prebuilt-base path for these 23 backends in CI, while local `make backends/<name>` (no extra args) keeps the from-source path. Mapping by (build-type, arch): - '' / amd64 -> base-grpc-amd64 - '' / arm64 -> base-grpc-arm64 - cublas-12 / amd64 -> base-grpc-cuda-12-amd64 - cublas-13 / amd64 -> base-grpc-cuda-13-amd64 - cublas-13 / arm64 -> base-grpc-cuda-13-arm64 - hipblas / amd64 -> base-grpc-rocm-amd64 - vulkan / amd64 -> base-grpc-vulkan-amd64 - vulkan / arm64 -> base-grpc-vulkan-arm64 - sycl_* / amd64 -> base-grpc-intel-amd64 - cublas-12 + JetPack r36.4.0 / arm64 -> base-grpc-l4t-cuda-12-arm64 Cold-build savings expected: ~25-35 min per variant (skips the gRPC compile + toolchain install that's now in the base). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci: add base-grpc-l4t-cuda-12-arm64 variant for legacy JetPack entries Two matrix entries (-nvidia-l4t-arm64-llama-cpp, -nvidia-l4t-arm64- turboquant) build against nvcr.io/nvidia/l4t-jetpack:r36.4.0 + CUDA 12 ARM64. They're distinct from -nvidia-l4t-cuda-13-arm64-* which use Ubuntu 24.04 + CUDA 13 sbsa. Add the missing JetPack-based variant to base-images.yml so those two entries' builder-base-image mapping in the previous commit resolves. Bootstrap order before merging this PR (re-run base-images.yml on this branch — 9 existing variants hit BuildKit cache, only the new l4t-cuda-12-arm64 builds cold): gh workflow run base-images.yml --ref ci/base-images-consumers Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci: extract base-builder install logic into .docker/install-base-deps.sh Pre-extraction, the apt + protoc + cmake + conditional CUDA/ROCm/Vulkan + gRPC install logic was duplicated across four files: - backend/Dockerfile.base-grpc-builder (CI prebuilt-base source of truth) - backend/Dockerfile.llama-cpp (builder-fromsource stage) - backend/Dockerfile.ik-llama-cpp (builder-fromsource stage) - backend/Dockerfile.turboquant (builder-fromsource stage) A bump to e.g. CUDA toolkit packages had to be made in 4 places, and drift between the prebuilt base and the variant-Dockerfile from-source path was a real concern (ik-llama-cpp's hipblas branch was already missing the rocBLAS Kernels echo that llama-cpp / turboquant / base-grpc-builder all had). Factor the install logic into a single .docker/install-base-deps.sh that reads its inputs from env vars and runs conditionally on BUILD_TYPE / CUDA_*_VERSION / TARGETARCH. Each Dockerfile now bind- mounts the script alongside .docker/apt-mirror.sh and invokes it from a single RUN step. The variant Dockerfiles' grpc-source stage is removed entirely — the script handles gRPC compile + install at /opt/grpc, and the builder-fromsource stage mirrors builder-prebuilt by copying /opt/grpc/. to /usr/local/. Result: - install-base-deps.sh: 244 lines (one source of truth) - Dockerfile.base-grpc-builder: 268 -> 98 lines - Dockerfile.llama-cpp: 361 -> 157 lines - Dockerfile.ik-llama-cpp: 348 -> 151 lines - Dockerfile.turboquant: 355 -> 154 lines - Total Dockerfile bytes: 1332 -> 560 lines (58% reduction) Bit-equivalence between prebuilt and from-source paths is now enforced by construction: both invoke the same script with the same inputs. A side-effect is that ik-llama-cpp now also gets the rocBLAS Kernels echo + clblas block parity it was previously missing. Includes the BUILD_TYPE=clblas branch (libclblast-dev) for parity even though no current CI matrix entry uses it. After this commit's force-push, base-images.yml needs to be redispatched on this branch — the Dockerfile.base-grpc-builder content shifts so the existing cache won't apply for the install layer (gRPC layer also rebuilds since it's now in the same RUN step). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> * ci(base-images): skip-drivers on JetPack l4t variant cuda-nvcc-12-0 isn't installable via apt on the JetPack r36.4.0 base image — JetPack ships CUDA preinstalled at /usr/local/cuda and its apt feed doesn't carry the cuda-nvcc-* packages from the public repositories. The original matrix entry for -nvidia-l4t-arm64-llama-cpp on master sets skip-drivers: 'true' for exactly this reason; the new base-grpc-l4t-cuda-12-arm64 base needs to match. Also forwards SKIP_DRIVERS as a build-arg from matrix into the build (was missing entirely before this commit). Caught by run 25612030775 — l4t-cuda-12-arm64 failed at: E: Package 'cuda-nvcc-12-0' has no installation candidate Assisted-by: Claude:claude-opus-4-7 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> |
||
|
Ettore Di Giacinto
|
acc5588d2c |
ci(darwin): force-link brew formulas after cache restore
Symptom: `ccache: command not found` in the Configure ccache step on runs that hit the brew cache. Root cause: actions/cache restores /opt/homebrew/Cellar/<formula> but NOT the bin symlinks at /opt/homebrew/bin/*. The subsequent `brew install` sees the Cellar entries present and decides "already installed" — without re-running the link step. So on cache-hit runs none of the cached formulas are actually on PATH. Fix: explicit `brew link --overwrite` for every formula we install, right after `brew install`. --overwrite tolerates leftover symlinks from a partial earlier install. The 2>/dev/null + || true keeps the step from failing if a formula is already correctly linked. Pre-existing flake; surfaces more often as Darwin matrix coverage grows after the llama-cpp-darwin consolidation in #9731. Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> |
||
|
LocalAI [bot]
|
28e29625a2 |
ci: add pre-built base-grpc-builder image infrastructure (PR 1/2) (#9737)
Introduces a parameterized Dockerfile.base-grpc-builder that produces
a fully-prepped builder base image (apt deps + protoc + cmake + gRPC
at /opt/grpc + conditional CUDA/ROCm/Vulkan toolchains) and a
base-images.yml workflow that builds + pushes 9 variants to
quay.io/go-skynet/ci-cache:base-grpc-*:
base-grpc-amd64 (Ubuntu 24.04, CPU-only)
base-grpc-arm64 (Ubuntu 24.04, CPU-only)
base-grpc-cuda-12-amd64 (Ubuntu 24.04 + CUDA 12.8)
base-grpc-cuda-13-amd64 (Ubuntu 22.04 + CUDA 13.0)
base-grpc-cuda-13-arm64 (Ubuntu 24.04 + CUDA 13.0 sbsa)
base-grpc-rocm-amd64 (rocm/dev-ubuntu-24.04:7.2.1 + hipblas)
base-grpc-vulkan-amd64 (Ubuntu 24.04 + Vulkan SDK 1.4.335)
base-grpc-vulkan-arm64 (Ubuntu 24.04 + Vulkan SDK ARM 1.4.335)
base-grpc-intel-amd64 (intel/oneapi-basekit:2025.3.2)
The variant Dockerfiles (Dockerfile.llama-cpp, ik-llama-cpp, turboquant)
are NOT touched in this PR. PR 2 will refactor them to FROM these
prebuilt bases. This PR is intentionally inert - landing it changes no
existing CI behavior. The base images don't exist on quay until
someone manually triggers the workflow.
Bootstrap after merge:
gh workflow run base-images.yml --ref master
Wait ~30 min for all 9 variants to push, then merge PR 2 (the
consumer-side refactor that uses BUILDER_BASE_IMAGE build-arg to
FROM these tags).
Triggers afterwards:
- Saturdays 05:00 UTC (cron) - picks up upstream security updates,
runs ~24h before the backend.yml Sunday cron so bases are fresh.
- workflow_dispatch - manual ad-hoc rebuild.
- master push touching Dockerfile.base-grpc-builder or this workflow.
Why split into two PRs: the variant Dockerfiles in PR 2 will FROM the
prebuilt bases and have no from-source fallback. Their CI builds fail
if the bases don't exist on quay yet. Landing infrastructure first +
manual bootstrap + then consumer refactor avoids a broken-master window.
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
|
||
|
Ettore Di Giacinto
|
31aa0582a5 |
ci(ik-llama-cpp,turboquant): add BuildKit ccache mount to compile steps
Mirror the ccache mount added to Dockerfile.llama-cpp in
|
||
|
Ettore Di Giacinto
|
3568b2819d |
fix(gallery): keep auto-upgrade off non-dev backends when -development is installed (#9736)
A `-development` backend variant (e.g. `cuda12-llama-cpp-development`)
shares its `alias` with the stable counterpart and is meant to be a
drop-in replacement via ListSystemBackends alias resolution. Two paths
in the auto-upgrade flow let the stable variant slip back in on top of
the user's explicit dev pick:
1. ListSystemBackends emits a synthetic alias row keyed by the alias
name that re-uses the chosen concrete's metadata pointer. In
distributed mode, the worker's handleBackendList serialised that
row over NATS as `{Name: <alias>, URI: <dev URI>, Digest: <dev>}`
— the frontend can't reconstruct the alias relationship, and the
wire-rebuilt row then carried `Metadata.Name = <alias>` and
resolved against an unrelated gallery entry on the next upgrade
check.
2. CheckUpgradesAgainst happily iterated the synthetic row in
single-node too. Today the duplicate gallery lookup is harmless
because both rows share the same `Metadata.Name`, but any gallery
change that gives a meta backend a version, or any concrete
sharing its alias with a dev counterpart, would surface a phantom
non-dev upgrade and auto-upgrade would install it — shadowing the
dev one through alias-token preference.
Two layered fixes:
- `core/services/worker/lifecycle.go` (`handleBackendList`): drop
rows where the map key differs from `b.Metadata.Name`. Concrete
and meta entries always have `key == Metadata.Name`; only synthetic
aliases violate it. Workers now report only what's actually on disk;
the per-node UI listing and CheckUpgrades both stop seeing phantoms.
- `core/gallery/upgrade.go` (`CheckUpgradesAgainst`): iterate by key,
skip rows where `key != Metadata.Name` (belt-and-suspenders for any
caller-supplied installed set), and apply the dev-aware rule —
build a set of installed `Metadata.Name`s and drop any non-dev
candidate `X` whose `X-<devSuffix>` counterpart is installed. Uses
the configured dev suffix from `getFallbackTagValues(systemState)`.
Manual `POST /api/backends/upgrade/<name>` is unaffected: it goes
straight through `bm.UpgradeBackend(name)` without consulting the
suppression list, so users who genuinely want the stable variant
upgraded can still trigger it explicitly.
Tests in core/gallery/upgrade_test.go cover three cases under
"CheckUpgradesAgainst (distributed)": dev-only installed → only the
dev surfaces; both variants installed → dev still wins; synthetic
alias row is ignored. Generic backend names are used to avoid the
capability filter dropping cuda-prefixed entries on a CPU-only host.
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
|
||
|
Ettore Di Giacinto
|
9228e5b412 |
ci(llama-cpp): add BuildKit ccache mount to the compile step
The big RUN at line 268 of Dockerfile.llama-cpp re-runs from scratch on every LLAMA_VERSION bump (or any LocalAI source change due to COPY . /LocalAI just before). For CUDA-13 specifically that compile recently hit the GHA 6h hard limit and failed: https://github.com/mudler/LocalAI/actions/runs/25598418931/job/75148244557 Add a BuildKit cache mount on /root/.ccache and thread ccache through CMake (CMAKE_C/CXX/CUDA_COMPILER_LAUNCHER) so most translation units hit cache when their preprocessed source is byte-identical to the previous build. The cache mount is exported to the registry as part of the existing cache-to: type=registry,mode=max in backend_build.yml, so it persists across runs. mount id is keyed on TARGETARCH + BUILD_TYPE so different variants don't thrash the same cache slot; sharing=locked serializes concurrent writes. Cold-build effect (first run after enable, or on LLAMA_VERSION bump that touches every TU): unchanged. Hot-build effect (subsequent runs with the same source, or LLAMA_VERSION bumps that touch a handful of files): ~5-15 min for the llama.cpp compile vs the previous 1-3h cold. For CUDA-13 specifically this should bring rebuilds well under the 6h GHA limit. Does NOT help the *first* post-bump build — that's still cold. For that, follow-up work would be: (a) trim CUDA_DOCKER_ARCH to modern GPUs only, (b) audit which CMake variants the published images actually need, (c) pre-built CUDA+gRPC base image. ccache package is already installed in the builder stage (line 90). Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Ettore Di Giacinto <mudler@localai.io> |
||
|
LocalAI [bot]
|
a91e718473 |
chore: ⬆️ Update ggml-org/llama.cpp to 00d56b11c3477b99bc18562dc1d1834f0d961778 (#9733)
⬆️ Update ggml-org/llama.cpp Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com> |