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 supported5 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [WIP][SPMD] Support spec decoding
- `[Core]` Added streaming support to `LLM` Class
- [Bug]: Guided decoding is broken because tokenizers can't be pickled
- [Bug]: online fp8 quantization with jais model got assert error due to cutlass_scaled_mm()
- [Bug]: fp8 performance is worse than fp16 when batch size is 1
- [Feature]: Context Parallelism
- [Usage]: Extremely slow inference with Llama 3.1 70b Instruct
- [Feature]: need a GB-based alternative for gpu_memory_utilization
- add causal parameter for flash attention
- [Bug]: error while attempting to bind on address ('0.0.0.0', 8000): address already in use
- Docs
- Python not yet supported