vllm
https://github.com/vllm-project/vllm
Python
A high-throughput and memory-efficient inference and serving engine for LLMs
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported3 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Feature]: load/unload API to run multiple LLMs in a single GPU instance
- [Hardware][Intel] fp8 kv cache support for CPU
- [Bug]: ray not work when tp>=2
- [Bug]: ailed to import from vllm._C with ImportError('/usr/local/lib/python3.8/dist-packages/vllm/_C.abi3.so: undefined symbol: _ZN5torch7LibraryC1ENS0_4KindESsSt8optionalIN3c1011DispatchKeyEEPKcj')
- [Usage]: OpenRLHF: How can I create a second NCCL Group in a vLLM v0.4.3+ Ray worker?
- [Bug]: Qwen/Qwen2-72B-Instruct 128k server down
- [Bug]: RuntimeError: out must have shape (total_q, num_heads, head_size_og)
- [Bug]: Loading Mixtral-8x22B-Instruct-v0.1-FP8 on 8xL40S causes a SIGSEGV
- [Bug]: Build from source but got unexpected symbol
- [Feature]: Allow user defined extra request args to be logged in OpenAI compatible server
- Docs
- Python not yet supported