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
- [Bug] NameError: name 'deep_gemm' is not defined in tf32_hc_prenorm_gemm when SGLANG_ENABLE_JIT_DEEPGEMM=0
- [sgl-model-gateway] fix tokenizer load blocking /health on tokio worker thread
- [AMD/ROCm] store_cache_4d Triton kernel fails to compile on ROCm 7.2.0 (TritonAMDGPUCanonicalizePointers)
- [TEST Dont Merge] test: cover batch input_ids routing in multi-tokenizer mode
- fix(eval): extract GSM8K answer from first #### marker, not last number
- [diffusion] feat: per-layer TP shard planner for DiT linears (--dit-tp-plan)
- fix(model_config): derive MLA scaling for MiniCPM3/DeepSeek-VL2/Kimi-VL and guard zero v_head_dim
- Bug Report: SGLang 0.5.14 Multi-Node GGUF MoE Deployment — 9 Bugs Fixed by a Finance Guy with an AI Agent
- [Feature] Add --config-format alias for --model-config-parser (Mistral vLLM parity, part 1 of #28020)
- fix: preserve resolved LongCat n-gram configuration
- Docs
- Python not yet supported