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 supported51 Subscribers
View all SubscribersAdd a CodeTriage badge to vllm
Help out
- Issues
- [Perf][Kernel] Adopt PTX 9.4 `ldmatrix.s8.s4` (hardware INT4βINT8 expanding load) in W4A8-INT8 paths
- [Feature]: Make PG_WAIT_TIMEOUT configurable via environment variable
- [CI] Prototype for Rust coverage
- [Bugfix][Tool Parser] Fix hunyuan_a13b/xlam streaming: argument named "name" spawns a phantom tool call
- [Feature] Support KV-cache events & KV-transfer for hybrid-attention models
- [Feature]: Real-time request-queue signal (NUM_REQUESTS_RUNNING/WAITING) for external routers
- [RFC]: Speculative recompute fallback for slow remote KV cache loads
- FlashInfer + spec-decode silently downgrades to PIECEWISE cudagraphs (β16% measured); warning understates the cost
- [Performance]: Dynamic speculative decoding (num_speculative_tokens_per_batch_size) causes catastrophic aggregate-throughput collapse under concurrency at the batch-size threshold (MTP, V1/PIECEWISE)
- [Bug]: MTP draft model ignores checkpoint per-layer quantization config (block_name_to_quantize mapped inconsistently vs target)
- Docs
- Python not yet supported