Open Issues Need Help
View All on GitHub Available unified memory in llama.cpp about 11 hours ago
good first issue
Add Pascal GPU (sm_61/sm_62) support to llama.cpp CUDA build about 1 month ago
good first issue
Docker model runner container does not sync changes to config.json; requires restart (engine only) about 1 month ago
good first issue
Docker Healthcheck failing about 1 month ago
good first issue
[Bug]vllm backend doesn't work for wsl2 ubuntu22.04 about 1 month ago
good first issue
Full support for v1/responses endpoint. about 1 month ago
good first issue
Unable to set vllm serve parameters using hf_overrides about 2 months ago
good first issue
NPU support about 2 months ago
good first issue
good first issue
docker model launch llmfit about 2 months ago
good first issue
Docker Sandboxes and DMR integration about 2 months ago
good first issue
Add support for Modelfile 2 months ago
good first issue
Merge the functionality of the existing model-runner and model-cli into a single dmr executable 2 months ago
good first issue
alt (option) + b/f doesn't work on iTerm2 on macOS 3 months ago
good first issue
good first issue
enhancement good first issue
Implement open prompt in system text editor 3 months ago
good first issue
The Rest of the OpenAI API support? 3 months ago
good first issue
e2e-test unreliable 3 months ago
good first issue
enhancement good first issue
AMD/ROCm implementation and testing (llama.cpp/vLLM) 5 months ago
good first issue
Support pushing to Hugging Face 5 months ago
good first issue
Hugging Face models produce non-deterministic OCI digests due to `time.Now()` in creation date 5 months ago
good first issue
good first issue
Model ID naming issues 5 months ago
good first issue
Model unavailable after sending first message 5 months ago
good first issue
Qwen3-vl and Qwen2.5-vl Video Mode 5 months ago
good first issue
"docker model run" optimization 5 months ago
good first issue
vLLM ROCm 5 months ago
good first issue
enhancement good first issue
good first issue
docker model show 5 months ago
good first issue
good first issue
Ray, Docker Model Runner, docker-compose integration 5 months ago
good first issue
Ray, Docker Model Runner, docker-compose 5 months ago
good first issue
Introduce log levels with slog 5 months ago
good first issue
Consistent UX for pull and push 6 months ago
good first issue
Pull vllm-compatible models from HuggingFace 6 months ago
good first issue
Enhancement: Expose richer model metadata in /engines/v1/models (similar to OpenRouter) 7 months ago
good first issue
Create "docker model bench MODEL" 7 months ago
good first issue
Unfork llama.cpp 7 months ago
good first issue
Implement and test Anthropic Messages API 7 months ago
good first issue
Feature: SGLang backend 7 months ago
good first issue
bug good first issue
Inconsistent units ("M", " MiB", "M", " M") 8 months ago
good first issue
Feature: MLX backend 9 months ago
enhancement good first issue
good first issue
Nvidia NIM support 9 months ago
good first issue nvidia
On Ctrl-C, we should just stop the bot from chatting, rather than exit the application altogether 9 months ago
good first issue
"docker model run --detach" 9 months ago
good first issue
Bind to interface 0.0.0.0 9 months ago
enhancement good first issue