mirror of
https://github.com/mudler/LocalAI.git
synced 2026-07-04 05:16:42 -04:00
eb32cd9073
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>
51 lines
2.1 KiB
Go
51 lines
2.1 KiB
Go
package localaitools
|
|
|
|
// Tool names exposed by the LocalAI Assistant MCP server. Use these
|
|
// constants — never bare strings — when registering tools, asserting the
|
|
// catalog in tests, or referencing tool names from other packages. The
|
|
// embedded skill prompts under prompts/ keep the bare strings because
|
|
// go:embed-ed markdown can't reference Go constants; TestPromptsContain
|
|
// SafetyAnchors guards that those strings stay aligned.
|
|
const (
|
|
// Read-only tools.
|
|
ToolGallerySearch = "gallery_search"
|
|
ToolListInstalledModels = "list_installed_models"
|
|
ToolListGalleries = "list_galleries"
|
|
ToolGetJobStatus = "get_job_status"
|
|
ToolGetModelConfig = "get_model_config"
|
|
ToolListBackends = "list_backends"
|
|
ToolListKnownBackends = "list_known_backends"
|
|
ToolSystemInfo = "system_info"
|
|
ToolListNodes = "list_nodes"
|
|
ToolVRAMEstimate = "vram_estimate"
|
|
ToolGetBranding = "get_branding"
|
|
ToolGetUsageStats = "get_usage_stats"
|
|
ToolGetPIIEvents = "get_pii_events"
|
|
ToolGetMiddlewareStatus = "get_middleware_status"
|
|
ToolGetRouterDecisions = "get_router_decisions"
|
|
|
|
// Mutating tools — guarded by Options.DisableMutating and the
|
|
// LLM-side safety prompt (see prompts/10_safety.md).
|
|
ToolInstallModel = "install_model"
|
|
ToolImportModelURI = "import_model_uri"
|
|
ToolDeleteModel = "delete_model"
|
|
ToolEditModelConfig = "edit_model_config"
|
|
ToolReloadModels = "reload_models"
|
|
ToolLoadModel = "load_model"
|
|
ToolInstallBackend = "install_backend"
|
|
ToolUpgradeBackend = "upgrade_backend"
|
|
ToolToggleModelState = "toggle_model_state"
|
|
ToolToggleModelPinned = "toggle_model_pinned"
|
|
ToolSetBranding = "set_branding"
|
|
ToolSetAlias = "set_alias"
|
|
|
|
// ToolListAliases is read-only but lives here so the alias tools stay
|
|
// grouped; the catalog tests assert its read-only placement.
|
|
ToolListAliases = "list_aliases"
|
|
)
|
|
|
|
// DefaultServerName is the MCP Implementation.Name surfaced when
|
|
// Options.ServerName is empty. Use the constant when you want a stable
|
|
// reference across packages (e.g. test fixtures, CLI defaults).
|
|
const DefaultServerName = "localai-admin"
|