Skip to content

Pull requests: sgl-project/sglang

New pull request New
2,535 Open 15,827 Closed
2,535 Open 15,827 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[PD] Add KV transfer metric fallback tests
#24446 opened May 5, 2026 by OCWC22 Loading…
3
[fix] /pause_generation and /continue_generation wrong for --tokenizer-worker-num > 1
#24445 opened May 5, 2026 by maocheng23 Contributor Loading…
4 tasks
1
5
deepseek_v2: route mega-MoE pre-dispatch through DeepGEMM + FP4 acts opt-in deepseek
#24444 opened May 5, 2026 by pranjalssh Contributor Loading…
3 tasks done
4
moe: add DeepEP V2 ElasticBuffer support to MoE token dispatcher
#24443 opened May 5, 2026 by dmvevents • Draft
4
test(prefill-delayer): pin offline-gen throughput test to triton backend
#24442 opened May 5, 2026 by YAMY1234 Contributor Loading…
2 tasks
7
[Docs] Add B200, GB200, GB300 NVIDIA hardware platform support for Kimi-K2.6 documentation Improvements or additions to documentation
#24441 opened May 5, 2026 by zijiexia Contributor • Draft
5 tasks
3
fix(req_pool): bump pool.size to match actual tensor row count after #24243 run-ci
#24439 opened May 5, 2026 by JustinTong0323 Collaborator Loading…
1 of 3 tasks
5
[Gemma 4] Adding MTP support
#24436 opened May 5, 2026 by kpham-sgl Collaborator Loading…
13
Update Qwen3-Coder docs_new NVIDIA guidance documentation Improvements or additions to documentation
#24435 opened May 5, 2026 by wenscarl Collaborator Loading…
2
[NemotronH] Fix expert scale weight loading
#24434 opened May 5, 2026 by chfeng-cs Loading…
5 tasks done
1
1
[diffusion] fix: fix diffusion FSDP sharding diffusion SGLang Diffusion run-ci
#24431 opened May 5, 2026 by mickqian Collaborator • Draft
3
[model-support] Add support for Bamba
#24430 opened May 5, 2026 by ppraneth Contributor Loading…
5 tasks
2
Support NemotronHPuzzleForCausalLM
#24429 opened May 5, 2026 by netanel-haber Contributor • Draft
2
[Fix] Stop resetting cache_hit_rate to 0 in report_decode_stats (#20451)
#24427 opened May 5, 2026 by yangyonggit Loading…
2 tasks done
1
[AMD] Add MiMo-V2.5-Pro in nightly tests for MI30x and MI35x amd run-ci
#24426 opened May 5, 2026 by michaelzhang-ai Collaborator Loading…
3
[Fix] Guard against None logprob fields when retracted request is streamed (#23154)
#24423 opened May 5, 2026 by yangyonggit Loading…
2 tasks done
1
[NPU] fix profiler on NPU
#24422 opened May 5, 2026 by zhaozx-cn Contributor Loading…
5 tasks done
3
[LoRA] Fix qkv_proj LoRA buffer sizing when tp_size > num_key_value_heads lora
#24420 opened May 5, 2026 by gh1595 • Draft
5 tasks
2
[codex] Add diffusion performance mode defaults diffusion SGLang Diffusion documentation Improvements or additions to documentation jit-kernel
#24419 opened May 5, 2026 by mickqian Collaborator Loading…
2
fix(gateway): use http1_only() on health check client model-gateway
#24418 opened May 5, 2026 by bhaktatejas922 Loading…
[PD] Fix KV transfer metrics run-ci
#24416 opened May 5, 2026 by cctry Collaborator Loading…
7
[Fix] Fix KV transfer metrics using wrong time window
#24415 opened May 5, 2026 by yangyonggit Loading…
7 tasks done
1
1
[vlm][pixtral] support precomputed embeddings + processor output Multi-modal multi-modal language model npu
#24412 opened May 5, 2026 by 1fanwang • Draft
5 tasks done
2
Feat/true on policy qwen moe deterministic Issues on deterministic inference/kernels documentation Improvements or additions to documentation Multi-modal multi-modal language model npu quant LLM Quantization
#24408 opened May 5, 2026 by maocheng23 Contributor • Draft
5 tasks
1
[Test] Add unit tests for srt/layers/utils/logprob.py
#24406 opened May 5, 2026 by dsuarez01 Loading…
9 of 11 tasks
1
ProTip! Filter pull requests by the default branch with base:main.