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 supported24 Subscribers
Add a CodeTriage badge to vllm
Help out
- Issues
- [Bug]: FlashInfer CUTLASS MoE backend causes CUDA illegal memory access on H100 during CUDA graph capture (Qwen3-Next-80B BF16)
- [Bug]: Eagle3 speculative decoding CUDA device-side assert crash with gpt-oss-120b under concurrent requests (TP=8, H20)
- [Feature]: Can you achieve this
- [RFC]: Selective KV Cache offload
- Use CU_MEMCPY_SRC_ACCESS_ORDER_ANY for batch KV cache swaps
- [Performance] DSV3.2 Indexer: Overlap indexer op || q_b_proj + MLA RoPE on separate CUDA streams
- [Performance] DSV3.2 Indexer: Overlap indexer k+w path || q path on separate CUDA streams
- WIP: SarvamMLAForCausalLM (begin work)
- WIP: Tarsier2ForConditionalGeneration (begin work)
- KV cache compression via E8 lattice VQ — 10-33x with PagedAttention integration
- Docs
- Python not yet supported