mudler/LocalAI

View on GitHub

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

Stars: 49.2k
Forks: 4.5k
Open beginner issues: 0
Indexed issues: 97
Avg merge: 1d 3m
Merged PRs (30d): 239

Dominant language: Go
License: MIT
Last GitHub push: Sep 19, 2026
Latest indexed: Sep 19, 2026
Contributing guide: Contributing guide
Code of conduct: No code of conduct
Beginner labels: No beginner labels indexed

Beginner friendly Everything

Open All

97 open issues indexed

localai cannot run the recommended model Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mudler/LocalAI#12081 · Opened Sep 16, 2026
Docs: gfx1151 env vars are documented as "set automatically in the ROCm/hipblas image" but the image sets none of them Open

Difficulty 2/5 1-3 hours Newbie friendliness 74/100

mudler/LocalAI#12071 · 1 comment · Opened Sep 15, 2026
Socket activation not working Open

bug unconfirmed

Difficulty 3/5 1-2 days Newbie friendliness 42/100

mudler/LocalAI#12024 · 1 comment · Opened Sep 13, 2026
Gallery model audio-cpp-sortformer-diarization fails to install: SHA mismatch, reproducible across retries Open

Difficulty 2/5 1-3 hours Newbie friendliness 76/100

mudler/LocalAI#11995 · 1 comment · Opened Sep 12, 2026
Agent pool Postgres DSN is logged in cleartext, including the password, while the auth DSN is redacted Open

Difficulty 2/5 1-3 hours Newbie friendliness 75/100

mudler/LocalAI#11991 · 1 comment · Opened Sep 12, 2026
long_term_memory: true without an embedding model crashes the entire server (unrecovered panic in saveCurrentConversation) Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 68/100

mudler/LocalAI#11975 · 1 comment · Opened Sep 11, 2026
docker-compose.yaml default model phi-2 no longer exists in the gallery Open

area/container bug confirmed

Difficulty 3/5 1-2 days Newbie friendliness 84/100

mudler/LocalAI#11974 · Opened Sep 11, 2026
docker-compose.yaml GPU example: capabilities [gpu, utility] omits compute, so libcuda.so.1 is not injected Open

Difficulty 2/5 1-3 hours Newbie friendliness 84/100

mudler/LocalAI#11973 · 1 comment · Opened Sep 11, 2026
vllm-omni backend (cuda13) fails to load: editable install points at build path, and unpinned vllm-omni main no longer matches backend.py Open

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mudler/LocalAI#11972 · 1 comment · Opened Sep 11, 2026
Expose per-model memory usage (the per-process VRAM walk already computes it, then discards the split) Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 68/100

mudler/LocalAI#11970 · 1 comment · Opened Sep 10, 2026
/api/ps and /api/tags report size and size_vram as a hardcoded 0 Open

area/api bug confirmed

Difficulty 3/5 1-2 days Newbie friendliness 68/100

mudler/LocalAI#11969 · 1 comment · Opened Sep 10, 2026
omp gets no response from localai Open

bug unconfirmed

Difficulty 3/5 1-2 days Newbie friendliness 42/100

mudler/LocalAI#11928 · 3 comments · Opened Sep 8, 2026
intel-sycl-f16-stablediffusion-ggml: SIGABRT during CGO call Open

area/backends area/container bug need-more-information os/gnu-Linux unconfirmed waiting-from-reporter

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mudler/LocalAI#11896 · 2 comments · Opened Sep 6, 2026
Quantization import always writes backend: llama-cpp, ignoring the backend that produced the model Open

Difficulty 2/5 1-3 hours Newbie friendliness 78/100

mudler/LocalAI#11875 · Opened Sep 4, 2026
Model Gallery sizing ("YOUR HOST") ignores registered Distributed Mode nodes Open

area/p2p area/webui bug need-more-information waiting-from-reporter

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mudler/LocalAI#11814 · 4 comments · Opened Sep 1, 2026
Add Embedding, Reranker (and optional OCR) model fields to Agent configuration for building RAG pipelines at the agent level Open

area/api area/webui enhancement

Difficulty 5/5 Over a week Newbie friendliness 42/100

mudler/LocalAI#11753 · Opened Aug 27, 2026
Upgrade backends fails if auto-upgrade is enabled Open

area/backends bug confirmed high prio

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#11720 · 1 comment · Opened Aug 25, 2026
Qwen3.5 llama-cpp tool calling fails when function has "strict": true Open

area/api area/llama.cpp bug confirmed

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mudler/LocalAI#11709 · Opened Aug 24, 2026
RWKV.cpp backend Open

area/backends enhancement model/rwkv new-backend

Difficulty 5/5 Over a week Newbie friendliness 35/100

mudler/LocalAI#11704 · Opened Aug 24, 2026
User-controlled Multi-Agent Conversation (Human-in-the-loop) Open

area/webui enhancement

Difficulty 5/5 Over a week Newbie friendliness 25/100

mudler/LocalAI#11703 · 1 comment · Opened Aug 24, 2026
Agent response disappears when navigating away via "Status" or switching chats in chat section Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#11702 · 1 comment · Opened Aug 24, 2026
Instella-MoE models in the LocalAI gallery is not supported by any backends Open

area/backends bug confirmed

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mudler/LocalAI#11681 · 1 comment · Opened Aug 23, 2026
Tool call arguments are empty on /responses but correct on /chat/completions (same model, same config) Open

area/api bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mudler/LocalAI#11635 · 1 comment · Opened Aug 20, 2026
audio-cpp using a fish-audio doesn't work with referance audio/text Open

area/backends bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mudler/LocalAI#11626 · 1 comment · Opened Aug 20, 2026
metal-whisper: ggml_metal_library_init fails to compile embedded Metal shader library on macOS 26.5.2 (Apple M4 Pro) — flash-attn kernel's threadgroup half4x4 array rejected Open

area/backends bug confirmed os/macOS upstream issue

Difficulty 5/5 Over a week Newbie friendliness 38/100

mudler/LocalAI#11624 · Opened Aug 20, 2026
Vision/multimodal silently dropped with use_tokenizer_template on the python backends (sglang, vllm): images reach the engine but the prompt carries no media placeholder Open

area/backends area/vllm bug confirmed

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mudler/LocalAI#11621 · 1 comment · Opened Aug 20, 2026
Voice Library Improvements Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 35/100

mudler/LocalAI#11617 · 1 comment · Opened Aug 20, 2026
TTS Voice Design WebUI Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 52/100

mudler/LocalAI#11609 · 2 comments · Opened Aug 19, 2026
API to list TTS voices Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mudler/LocalAI#11604 · 1 comment · Opened Aug 19, 2026
cuda13-llama-cpp backend crashes loading gemma4 GGUF: "wrong number of tensors; expected 2131, got 720" Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#11593 · 1 comment · Opened Aug 19, 2026
Fish Speech backend using ROCM doesn't work Open

bug unconfirmed

Difficulty 3/5 1-2 days Newbie friendliness 48/100

mudler/LocalAI#11579 · 1 comment · Opened Aug 18, 2026
ROCM audio-cpp image Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mudler/LocalAI#11577 · 1 comment · Opened Aug 18, 2026
voice-detect and face-detect backends SIGSEGV on ROCm (gfx1100) — null pointer in purego.RegisterFunc call Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mudler/LocalAI#11562 · 3 comments · Opened Aug 17, 2026
stablediffusion-ggml backend segfaults instead of returning an error when a Metal pipeline fails to compile, and its stderr never reaches the LocalAI log Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mudler/LocalAI#11529 · 1 comment · Opened Aug 15, 2026
whisperx doesn't return speaker when diarization is requested Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mudler/LocalAI#11520 · 1 comment · Opened Aug 14, 2026
No Mac os version 4.8.0 or 4.8.1 . Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#11410 · 1 comment · Opened Aug 7, 2026
fish-speech: make compile:true usable on Blackwell sm_121 by honouring the CUDA toolkit's ptxas Open

enhancement

Difficulty 2/5 1-3 hours Newbie friendliness 76/100

mudler/LocalAI#11348 · 1 comment · Opened Aug 4, 2026
fish-speech-s2-pro on localai/localai:latest-nvidia-l4t-arm64-cuda-13 (DGX Spark) Open

bug unconfirmed

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#11344 · 1 comment · Opened Aug 4, 2026
WebU(traces): Traces keep on scrolling after opening Open

bug unconfirmed

Difficulty 3/5 1-2 days Newbie friendliness 52/100

mudler/LocalAI#11277 · 1 comment · Opened Aug 1, 2026
add trellis.2 backend back 3d model generation Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 35/100

mudler/LocalAI#11184 · 1 comment · Opened Jul 29, 2026
Streaming: all deltas labeled 'reasoning' when the rendered prompt contains a (pre-closed) thinking marker — non-streaming parses the same output correctly Open

area/api area/llama.cpp bug unconfirmed

Difficulty 3/5 1-2 days Newbie friendliness 52/100

mudler/LocalAI#11135 · 1 comment · Opened Jul 26, 2026
`pinned: true` model is freed after every request — idle-unload exclusion (#9305) ineffective on v4.7.1 Open

Difficulty 4/5 3-5 days Newbie friendliness 52/100

mudler/LocalAI#11101 · 2 comments · Opened Jul 24, 2026
Support B70 openAPI Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#11077 · 3 comments · Opened Jul 23, 2026
Responses Streaming SSE Keepalive Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 55/100

mudler/LocalAI#11076 · 1 comment · Opened Jul 23, 2026
LocalAI QA: Make tests go fast Open

area/test enhancement

mudler/LocalAI#11048 · 1 comment · 1 assignee · Opened Jul 22, 2026
Feature request: official gfx1151 (Strix Halo / RDNA3.5) backend builds on ROCm 7.14 Open

area/backends area/ci enhancement

Difficulty 5/5 Over a week Newbie friendliness 45/100

mudler/LocalAI#10883 · 4 comments · Opened Jul 17, 2026
The vulnerability I submitted according to your security policy has not received a response for a long time Open

bug high prio unconfirmed

Difficulty 5/5 Over a week Newbie friendliness 15/100

mudler/LocalAI#10751 · 2 comments · Opened Jul 9, 2026
[bug] Image upgrade doesn't detect stale backend venv → cryptic 'Engine core init failed'; install.sh needs uv but nvidia-l4t image has neither uv nor pip Open

area/backends bug os/gnu-Linux

Difficulty 4/5 3-5 days Newbie friendliness 48/100

mudler/LocalAI#10720 · 3 comments · Opened Jul 7, 2026
qwythos-9b-claude-mythos-5-1m breaks on the second message Open

bug confirmed high prio

Difficulty 4/5 3-5 days Newbie friendliness 45/100

mudler/LocalAI#10659 · 4 comments · Opened Jul 3, 2026
Dockerised LocalAI on a system with integrated AMD GPU and shared memory cannot load any Stable diffusion or Diffusers model Open

area/backends blocked bug confirmed os/gnu-Linux

Difficulty 4/5 3-5 days Newbie friendliness 35/100

mudler/LocalAI#10636 · 3 comments · Opened Jul 1, 2026

mudler/LocalAI

Get new issues in your inbox