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
- [LoRA] MoE-LoRA CUDA-graph fixes under DP attention + RL adapter-reload fixes
- tmp
- [Fix] Simple fix: pass tokenizer stop tokens to XGrammar so constrained requests can terminate
- [1/N] [spec] align draft-extend graph-replay buckets with can_run_graph
- [Docker] Bump DEEPEP_COMMIT past hidden-dim-3072 support (pre-EPv2 tip)
- [DeepSeek] Eliminate per-step elementwise kernels on the BF16 MLA-absorb path (follow-up to #31604)
- [DeepSeek] Add SGLANG_MLA_ABSORB_FP8: per-tensor FP8 MLA-absorb weights (follow-up to #31604)
- [trtllm_mla] Fuse cuda-graph metadata rebuild into one triton kernel
- Avoid DSA MQA chunking caused by cached CUDA memory
- [DSA] Handle partial-DP padding in Decode page-table transform
- Docs
- Python not yet supported