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
- [Installation]: Failed to build punica
- [Misc] Adding Speculative decoding to Throughput Benchmarking script
- [Feature]: Add efficient interface for evaluating probabilities of fixed prompt-completion pairs
- [BugFix]Fix the problem that StopChecker assumes a single token produ…
- [Bug]: high gpu_memory_utilization with 'OOM' and low gpu_memory_utilization with 'No available memory for the cache blocks'
- [Bug]: speculative decoding with max-num-seqs <= 2 * num-speculative-tokens
- [Feature]: Support selecting chat template
- Debug the optimal upper-bound performance for swapping (0-cost swapping).
- How to use vllm to compute ppl score for input text?
- Guidance on how many requests can be processed at a time?
- Docs
- Python not yet supported