Skip to content

Pull requests: sgl-project/sglang

New pull request New
146 Open 2,570 Closed
146 Open 2,570 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

remove moe_align vllm dep
#4249 opened Mar 10, 2025 by sleepcoo Loading…
[Feature] Support EAGLE 3
#4247 opened Mar 10, 2025 by chromecast56 Loading…
6 tasks
2
[Feature] Support Tensor Parallelism and Weight Slicing for Lora
#4239 opened Mar 9, 2025 by aoshen524 Loading…
2 of 3 tasks
Integrate deepEP into SGLang high priority
#4232 opened Mar 9, 2025 by liz-badada • Draft
6 tasks
5
fix per_token_group_quant_fp8 illegal memory when num_groups % 16 != 0
#4231 opened Mar 9, 2025 by BBuf Loading…
5
refactor: move image processors to individual files
#4229 opened Mar 9, 2025 by mickqian Loading…
1 of 6 tasks
1
[Feature] Prefill assistant response - add continue_final_message parameter
#4226 opened Mar 9, 2025 by adarshxs Loading…
3 tasks done
1
Add H20 tuning configs support DeepSeek V3/R1 INT8(block-wise)
#4220 opened Mar 9, 2025 by Ximingwang-09 Loading…
4 of 6 tasks
Remove vllm ops scaled fp8 quant and accelerate per token quant by 25-30%
#4215 opened Mar 8, 2025 by hebiao064 Loading…
3 of 6 tasks
[Fix] Check the device backend before calling empty_cache function
#4212 opened Mar 8, 2025 by cboss6 Loading…
1 of 6 tasks
Added example for multimodal embedding
#4206 opened Mar 8, 2025 by simveit Loading…
3
Statistical Analysis of the Output Stability of the Deepseek Model
#4202 opened Mar 8, 2025 by tanzelin430 • Draft
2 of 6 tasks
1
linear support deepgemm high priority
#4199 opened Mar 8, 2025 by sleepcoo • Draft
Fix MoE quant args
#4190 opened Mar 8, 2025 by Edenzzzz Loading…
6 tasks
[ROCm/Draft/No-Merge]: Flex Attention Enablement amd collaboration documentation Improvements or additions to documentation
#4172 opened Mar 7, 2025 by HaiShaw • Draft
6 tasks
DeepGemm integrate to sgl-kernel high priority
#4165 opened Mar 7, 2025 by laixinn Loading…
6 tasks done
@zhyncs
13
[ROCm] Enable silu_and_mul, gelu_and_mul, gelu_tanh_and_mul in amd platform
#4150 opened Mar 6, 2025 by yiakwy-xpu-ml-framework-team Loading…
6 tasks
fix the input_ids is None error
#4144 opened Mar 6, 2025 by Young1993 Loading…
6 tasks
1
add --served-model-name arg for bench_serving
#4141 opened Mar 6, 2025 by gujingit Loading…
6 tasks
1
[QUANT] Support DeepSeek-V3 gptq
#4139 opened Mar 6, 2025 by rainkert Loading…
@hnyls2002
2
Add A800 tuning configs support DeepSeek V3/R1 BF16 and INT8(block-wise)
#4136 opened Mar 6, 2025 by lambert0312 Loading…
1 of 6 tasks
@ispobock
3
Add test for Radix cache variants
#4125 opened Mar 6, 2025 by Edenzzzz Loading…
6 tasks
7
Support cuda graph for LoRA
#4115 opened Mar 6, 2025 by Qiaolin-Yu • Draft
1 of 6 tasks
1
Add awq dequantize kernel to sgl with 1x to 3x speedup
#4104 opened Mar 5, 2025 by zcnrex Loading…
6 tasks
7
Enable the native path of DeepSeek
#4086 opened Mar 5, 2025 by airMeng Loading…
2 of 6 tasks
51
ProTip! What’s not been updated in a month: updated:<2025-02-09.