Morph MORPH
SIGN IN SIGN UP
mudler / LocalAI UNCLAIMED

LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.

0 0 66 Go

COMMITS

/ core/http/endpoints/openai/chat.go
master
May 25, 2026
L
fix(reasoning): stop <think> leaking into content when autoparser is in pure-content mode (#9991)
LocalAI [bot] committed
1c6c3ad
R
feat(middleware): Model routing, PII filtering, Cloud model proxies (#9802)
Richard Palethorpe committed
6a80e23
May 22, 2026
L
fix(openai): stream usage non-zero when tools are enabled (#9941)
LocalAI [bot] committed
0b2ae3c
May 14, 2026
L
fix(streaming): comply with OpenAI usage / stream_options spec (#9815)
LocalAI [bot] committed
8af963b
May 8, 2026
R
chore: Security hardening (#9719)
Richard Palethorpe committed
670259c
April 28, 2026
E
feat: localai assistant chat modality (#9602)
Ettore Di Giacinto committed
bcef72b
April 21, 2026
E
fix(streaming): dedupe content, recover reasoning, unique tool_call IDs in deferred flush (#9470)
Ettore Di Giacinto committed
7d8c1d5
April 9, 2026
E
fix(streaming): deduplicate tool call emissions during streaming (#9292)
Ettore Di Giacinto committed
e1a6010
E
fix: thinking models with tools returning empty content (reasoning-only retry loop) (#9290)
Ettore Di Giacinto committed
13a6ed7
April 6, 2026
E
fix(chat): do not retry if we had chatdeltas or tooldeltas from backend (#9244)
Ettore Di Giacinto committed
773489e
April 5, 2026
E
fix(autoparser): correctly pass by logprobs (#9239)
Ettore Di Giacinto committed
232e324
April 4, 2026
E
fix(reasoning): suppress partial tag tokens during autoparser warm-up
Ettore Di Giacinto committed
53deeb1
E
fix(reasoning): warm-up
Ettore Di Giacinto committed
c5a840f
E
fix(reasoning): accumulate and strip reasoning tags from autoparser results (#9227)
Ettore Di Giacinto committed
6d9d77d
R
feat(api): Allow coding agents to interactively discover how to control and configure LocalAI (#9084)
Richard Palethorpe committed
557d0f0
E
feat(autoparser): prefer chat deltas from backends when emitted (#9224)
Ettore Di Giacinto committed
716ddd6
March 29, 2026
E
feat: add distributed mode (#9124)
Ettore Di Giacinto committed
59108fb
March 21, 2026
E
feat: inferencing default, automatic tool parsing fallback and wire min_p (#9092)
Ettore Di Giacinto committed
031a36c
March 20, 2026
R
feat(ui, openai): Structured errors and link to traces in error toast (#9068)
Richard Palethorpe committed
8cd3f9f
March 16, 2026
E
chore: refactor endpoints to use same inferencing path, add automatic retrial mechanism in case of errors (#9029)
Ettore Di Giacinto committed
ee96e5e
March 15, 2026
E
feat: support streaming mode for tool calls in agent mode, fix interleaved thinking stream (#9023)
Ettore Di Giacinto committed
5fd4239
March 11, 2026
E
feat(ui): MCP Apps, mcp streaming and client-side support (#8947)
Ettore Di Giacinto committed
8818452
March 8, 2026
E
feat(functions): add peg-based parsing and allow backends to return tool calls directly (#8838)
Ettore Di Giacinto committed
b2f81bf
March 1, 2026
E
fix: retry when LLM returns empty messages (#8704)
Ettore Di Giacinto committed
c7c4a20
February 2, 2026
A
fix(api): Add missing field in initial OpenAI streaming response (#8341)
Alex O'Connell committed
b7585ca
January 20, 2026
E
feat(openresponses): Support reasoning blocks (#8133)
Ettore Di Giacinto committed
c491c6c
E
fix(reasoning): support models with reasoning without starting thinking tag (#8132)
Ettore Di Giacinto committed
34e054f
January 10, 2026
E
feat(api): support 'reasoning' api field (#7959)
Ettore Di Giacinto committed
c88074a
January 5, 2026
E
feat(function): Add tool streaming, XML Tool Call Parsing Support (#7865)
Ettore Di Giacinto committed
21c84f4
December 21, 2025
E
chore(refactor): move logging to common package based on slog (#7668)
Ettore Di Giacinto committed
c37785b
November 16, 2025
E
feat: add support to logitbias and logprobs (#7283)
Ettore Di Giacinto committed
d7f9f3a
November 14, 2025
E
feat: migrate to echo and enable cancellation of non-streaming requests (#7270)
Ettore Di Giacinto committed
1cdcaf0
E
chore(deps): bump llama.cpp to `c4abcb2457217198efdd67d02675f5fddb7071c2` (#7266)
Ettore Di Giacinto committed
7129409
November 9, 2025
C
fix(api): SSE streaming format to comply with specification (#7182)
Copilot committed
34bc1bd
E
feat: respect context and add request cancellation (#7187)
Ettore Di Giacinto committed
679d43c
November 7, 2025
E
feat(llama.cpp): consolidate options and respect tokenizer template when enabled (#7120)
Ettore Di Giacinto committed
02cc8cb
October 8, 2025
E
fix(grammars): handle empty parameters on object types (#6409)
Ettore Di Giacinto committed
df46a43
October 5, 2025
E
fix: add strict mode check for no action function (#6294)
Ettore Di Giacinto committed
fc8d5c9
September 11, 2025
M
fix(chat): use proper finish_reason for tool/function calling (#6243)
Mário Freitas committed
2c0b9c6
August 28, 2025
E
chore: stream errors while streaming SSE (#6160)
Ettore Di Giacinto committed
ad99399
August 14, 2025
E
feat(backends): add system backend, refactor (#6059)
Ettore Di Giacinto committed
089efe0
August 6, 2025
E
feat: add reasoning effort and metadata to template (#5981)
Ettore Di Giacinto committed
b9a25b1
July 24, 2025
D
fix: untangle pkg and core (#5896)
Dave committed
b3c2a3c
June 29, 2025
E
fix(gallery): automatically install model from name (#5757)
Ettore Di Giacinto committed
33f9ee0
February 10, 2025
D
feat: Centralized Request Processing middleware (#3847)
Dave committed
3cddf24
February 6, 2025
E
fix(openai): consistently return stop reason (#4771)
Ettore Di Giacinto committed
8d45670
January 18, 2025
M
chore(docs): extra-Usage and Machine-Tag docs (#4627)
mintyleaf committed
96306a3
January 17, 2025
M
feat: add machine tag and inference timings (#4577)
mintyleaf committed
96f8ec0
December 8, 2024
E
Revert "feat: include tokens usage for streamed output" (#4336)
Ettore Di Giacinto committed
f943c4b
E
feat(template): read jinja templates from gguf files (#4332)
Ettore Di Giacinto committed
cea5a0e