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
- Max throughout for llama3.2 1B model
- [Usage]: How to improve throughput with multi-card inference?
- [Bug]: lora is not loaded using `num_scheduler_steps=8`
- [Usage]: Are prompts processed sequentially?
- [Feature]: Multiple Secret Keys
- [Bug]: hung when start openai api server with multiple gpu in one node.
- [Feature]: Refactor Logging Part and Improve Logging For Embedding Models
- [Bug]: An EXTREMELY WEIRD bug when I import evaluate before vllm
- [Core] Support offloading KV cache to CPU
- [Bug]: "gettid" was not declared error when build from source for cpu with version after v0.6.1
- Docs
- Python not yet supported