sglang
https://github.com/sgl-project/sglang
Python
SGLang is a high-performance serving framework for large language models and multimodal models.
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 supported1 Subscribers
Add a CodeTriage badge to sglang
Help out
- Issues
- [sglang-miles] Allocate DSA cuda-graph page_table and flashmla_metadata in TMS cuda_graph region
- test: cover PP-local KV buffer shape
- [router] Return 504 Gateway Timeout, not 502, on upstream request timeout
- [npu] acc fix for Intern, a fast tokenizer via slow to fast conversion (e.g. InternLM2Converter) when tokenizer.json is missing.
- [Spec] Size shared logits buffer for adaptive candidates
- fix(server_args): normalize log level so uvicorn accepts it
- remove index_head_dim from OOT mla token_to_kv_pool
- [npu] [bugfix] Add usage constraints for fused_rope_qk_mqa fusion operator
- refactor _biased_grouped_topk_postprocess to _postprocess_topk_ids_cuda
- [BUG] Fix HiSparse preallocation for hybrid SWA decode
- Docs
- Python not yet supported