Issues
Closed
Eval bug: Model loading fails on Windows when path contains non-ASCII (Cyrillic) characters
bughelp wanted
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
8 comments0 reactions0 assignees
Closed
Feature Request: support Qwen3-omni, Qwen3-TTS and MiniCPM-o
Review Complexity : Highenhancementgood first issuehelp wantedmtmdtts
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
14 comments29 reactions0 assignees
Closed
Feature Request: Mapping model name to LoRA config
enhancementgood first issueserver
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
6 comments2 reactions0 assignees
Closed
examples : add configuration presets
documentationenhancementexamplesgood first issuehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
9 comments3 reactions0 assignees
Closed
Feature Request: Allow Filtering LLama Server Response Fields
enhancementgood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
12 comments0 reactions0 assignees
Closed
llama : add CLI assistant
enhancementgood first issueroadmap
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
5 comments4 reactions0 assignees
Closed
Bug: Failed to run qwen2-57b-a14b-instruct-fp16.
buggood first issuehigh severity
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
4 comments0 reactions0 assignees
Closed
Improve `cvector-generator`
enhancementgood first issuehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
34 comments1 reaction0 assignees
Closed
ggml : add WebGPU backend
help wantedresearch 🔬roadmap
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
28 comments23 reactions0 assignees
Closed
Llava functions compiled as extern "C" throw exceptions
buggood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
1 comment0 reactions0 assignees
Closed
server: avoid full prompt eval when 'prompt >= ctx'
enhancementgood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
3 comments0 reactions0 assignees
Closed
Support for Phi-3 models
good first issuemodel
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
84 comments61 reactions0 assignees
Closed
common: download from URL, improve parallel download progress status
enhancementgood first issuehelp wantedsplit
Why recommendedHas a beginner-friendly label · Repository active this month
Has a beginner-friendly labelRepository active this monthContributing guide available
7 comments0 reactions1 assignee
Closed
server: support control vectors
enhancementgood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
1 comment4 reactions0 assignees
Closed
server: exit failure if `--embedding` is set with an incoherent `--ubatch-size`
enhancementgood first issuehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
8 comments1 reaction0 assignees
Closed
llama : refactor the llm.build_xxx functions
good first issuerefactoringroadmap
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
3 comments3 reactions0 assignees
Closed
in situ auto-Frankenmerges
enhancementgood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
11 comments2 reactions0 assignees
Closed
convert.py incorrectly detects LLaMAv1 65B as a LLaMAv2 model
good first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
3 comments1 reaction0 assignees
Closed
llama : combined beam search + grammar sampling strategy
generation qualitygood first issueresearch 🔬roadmap
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
14 comments20 reactions0 assignees
Closed
feature request - disabling tokenizer in conversion / inference
enhancementgood first issuehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly labelRepository active this monthContributing guide available
7 comments1 reaction0 assignees