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 supported61 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Quantization][Perf] FP8-input consumer for block-FP8 linear on sm90
- [AuxOutput] use extensible token-aligned data format for mooncake data
- [Bugfix][PCP] Run the MTP draft prefill eagerly under prefill context parallel
- [Perf][MRV2] Compute text-only M-RoPE positions on the GPU
- [Perf][MRV2] Stage write contents in typed arrays
- [Perf][MRV2] Upload batch indices with one UVA kernel
- [Perf][MRV2] Opt-in suffix-only token staging for new requests
- [Profiler] Record the first N profiled iterations with stacks and shapes
- [Profiler] Record the graph warm-up next to its capture
- [RL] Support per-parameter destination layouts for NCCL M2N
- Docs
- Python not yet supported