Pull requests: sgl-project/sglang

Pull requests list

Revert "Set csgmv as default lora backend. (#11488)" high priority run-ci

#11735 by zhyncs was merged Oct 17, 2025

4 tasks

@lifuhuang @zhyncs @Fridge003

Cleaning indexer for DeepSeek V3.2 run-ci

#11682 by Fridge003 was merged Oct 17, 2025

4 tasks

@zhyncs

feat: add add_chunked_prefix_cache_attention_backend run-ci

#11636 by zhyncs was merged Oct 15, 2025

4 tasks

@zhyncs

Fix DeepSeek-v3.2 default config (ValueError: not enough values to unpack (expected 4, got 3)) high priority run-ci

#11557 by trevor-m was merged Oct 13, 2025

4 tasks

@zhyncs @Qiaolin-Yu

Update DeepSeek-R1-FP4 default config on blackwell high priority run-ci

#11512 by Qiaolin-Yu was merged Oct 13, 2025

4 tasks

@zhyncs

chore: bump sgl-kernel version to 0.3.16 run-ci

#11476 by sglang-bot was merged Oct 12, 2025

@zhyncs

Add metrics for speculative decoding (acceptance rate, average acceptance length) high priority run-ci

#11441 by scottjlee was merged Oct 13, 2025

4 tasks

@zhyncs @ShangmingCai

move eagle draft post process to cuda graph high priority run-ci

#11434 by cicirori was merged Oct 14, 2025

@merrymercy @zhyncs @hnyls2002

Beta spec-overlap for EAGLE high priority run-ci

#11398 by hnyls2002 was merged Oct 12, 2025

@cicirori @zhyncs

chore: upgrade flashinfer 0.4.0 high priority run-ci

#11364 by zhyncs was merged Oct 9, 2025

4 tasks

@zhyncs

fix: fix revision for sgl-flash-attn in sgl-kernel run-ci

#11327 by mickqian was merged Oct 8, 2025

4 tasks

@zhyncs

chore: bump SGLang version to 0.5.3.post1 run-ci

#11324 by sglang-bot was merged Oct 9, 2025

@zhyncs

chore: bump SGLang version to 0.5.3 high priority run-ci

#11263 by sglang-bot was merged Oct 6, 2025

@merrymercy @ispobock @zhyncs @Fridge003

Fix LoRA support for multimodal models (VLMs) by implementing a consistent pattern for skipping vision components high priority run-ci

#11261 by ConnorLi96 was merged Oct 7, 2025

1 of 4 tasks

@zhyncs

chore: bump SGLang version to 0.5.3rc2 high priority run-ci

#11259 by sglang-bot was merged Oct 6, 2025

@zhyncs

Fix DeepSeek chunked prefill memory issue run-ci

#11149 by fzyzcjy was merged Oct 2, 2025

4 tasks

@zhyncs

Add metrics for speculative decoding (acceptance rate, average acceptance length) high priority run-ci

#11144 by scottjlee was merged Oct 10, 2025

4 tasks

@merrymercy @ispobock @zhyncs @hnyls2002 @ShangmingCai

Allow use of TRTLLM_MHA backend for hybrid attention on Blackwell high priority run-ci

#11138 by DomBrown was merged Oct 2, 2025

4 tasks done

@lifuhuang @zhyncs @Fridge003

chore: upgrade sgl-kernel 0.3.14 high priority run-ci

#11107 by zhyncs was closed Oct 1, 2025

4 tasks

@FlamingoPg @zhyncs

chore: bump sgl-kernel v0.3.14 high priority run-ci

#11067 by FlamingoPg was merged Sep 30, 2025

4 tasks

@zhyncs

chore: upgrade sgl-kernel 0.3.13 run-ci

#11056 by zhyncs was merged Sep 29, 2025

4 tasks

@zhyncs

Create two new GH workflows to automatically bump SGLang and Kernel version high priority run-ci

#10996 by Kangyan-Zhou was merged Oct 6, 2025

@zhyncs

Fix profiling tools run-ci

#10977 by Qiaolin-Yu was closed Sep 28, 2025

4 tasks

@zhyncs

[1/2] Support FA4 for MHA Prefill in sgl-kernel high priority ready-to-merge run-ci

#10940 by lifuhuang was merged Sep 29, 2025

@zhyncs

[2/2] Support MHA prefill with FlashAttention 4. high priority run-ci

#10937 by lifuhuang was merged Oct 8, 2025

4 tasks

@zhyncs

ProTip! no:milestone will show everything without a milestone.

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Pull requests: sgl-project/sglang

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Pull requests list