Skip to content

Pull requests: sgl-project/sglang

New pull request New
2,850 Open 18,729 Closed
2,850 Open 18,729 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

[session] Fix reaper closing sessions mid-decode; forward session_params in /v1/completions
#28140 opened Jun 13, 2026 by discobot Loading…
3 of 5 tasks
1
5
[CI] Kernel benchmark regression gate (draft / RFC) documentation Improvements or additions to documentation sgl-kernel
#28138 opened Jun 13, 2026 by BBuf Collaborator • Draft
4 tasks
5
[Kernel] Add SM120/SM121 dispatch for int8_scaled_mm sgl-kernel
#28137 opened Jun 13, 2026 by waynehacking8 Loading…
3 tasks done
3
[Core] Reduce RoPE cache size for shorter context length
#28136 opened Jun 13, 2026 by labAxiaoming Loading…
5 tasks
2
fix(openai): reject request-supplied chat_template by default
#28135 opened Jun 13, 2026 by Sunt-ing Contributor Loading…
5 tasks done
[Spec] Clear dead DRAFT_EXTEND objects left after EAGLE v1 removal bypass-fastfail run-ci run-ci-extra
#28133 opened Jun 13, 2026 by ch-wan Collaborator Loading…
2
fix(quant): support block FP8 strategy for CompressedTensorsW8A16Fp8 quant LLM Quantization
#28132 opened Jun 13, 2026 by Sunt-ing Contributor Loading…
4 tasks done
3
fix(multimodal): return 400 for corrupt image inputs
#28131 opened Jun 13, 2026 by Sunt-ing Contributor Loading…
3 tasks done
3
fix(quant): dispatch compressed-tensors ParallelLMHead quant LLM Quantization
#28130 opened Jun 13, 2026 by Sunt-ing Contributor Loading…
4 tasks done
3
[Spec] Remove deprecated EAGLE v1 DRAFT_EXTEND forward mode blackwell SM100/SM120 bypass-fastfail deepseek npu run-ci run-ci-extra speculative-decoding
#28129 opened Jun 13, 2026 by ch-wan Collaborator Loading…
4
feat(cookbook): generic config-declared flagSelects playground axis documentation Improvements or additions to documentation
#28128 opened Jun 13, 2026 by zijiexia Collaborator Loading…
3
[diffusion] Improve server warmup coverage diffusion SGLang Diffusion run-ci
#28127 opened Jun 13, 2026 by mickqian Collaborator Loading…
3
[NPU] add device print
#28126 opened Jun 13, 2026 by zhaozx-cn Contributor Loading…
5 tasks
1
feat(moe): add SM120/SM121 dispatch for fp8_blockwise_scaled_grouped_mm sgl-kernel
#28125 opened Jun 13, 2026 by waynehacking8 Loading…
3 tasks done
1
[AMD] GQA-grouped split-K verify kernel for EAGLE3/MTP extend attention
#28124 opened Jun 13, 2026 by JohnQinAMD Loading…
5 tasks
1
[diffusion] test: tighten perf baselines diffusion SGLang Diffusion run-ci
#28123 opened Jun 13, 2026 by mickqian Collaborator Loading…
3
[MLX] Add Metal profiling hooks to bench_offline_throughput and server profiler
#28122 opened Jun 13, 2026 by LijuanTang94 Contributor Loading…
3
[HiSparse] fix: populate req_pool_indices_cpu in staging-to-decode batch run-ci
#28121 opened Jun 13, 2026 by alphabetc1 Collaborator Loading…
@xiezhq-hermann
11
【bugfix】The NPU's forward_dsa_prepare_npu also needs special handling for is_nextn deepseek npu
#28118 opened Jun 13, 2026 by littleyellowbicycle Contributor Loading…
4 tasks
1
[Need to test on Node] Fix Responses API review follow-ups from #25881
#28115 opened Jun 13, 2026 by alexnails Collaborator • Draft
1
fix: preserve GLM string ids with underscores
#28114 opened Jun 13, 2026 by he-yufeng Contributor Loading…
1
1
[Platform] Route pin memory availability through current_platform documentation Improvements or additions to documentation
#28113 opened Jun 13, 2026 by N3u0ns Loading…
5 tasks done
@alexnails
4
[LoRA] Support DSA indexer LoRA targets for GLM-5.1 / DeepSeek-V3.2-family models lora run-ci
#28110 opened Jun 13, 2026 by jybsuper Collaborator Loading…
2
Relax apache-tvm-ffi dependency constraints dependencies Pull requests that update a dependency file
#28109 opened Jun 13, 2026 by sladyn98 Loading…
5 tasks done
2
[attn backend] Make seq_lens_cpu optional in trtllm_mha backend blackwell SM100/SM120
#28106 opened Jun 12, 2026 by JonnyKong Contributor Loading…
4 of 5 tasks
2
ProTip! Add no:assignee to see everything that’s not assigned.