-
Notifications
You must be signed in to change notification settings - Fork 3.5k
Pull requests: sgl-project/sglang
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
diffusion: improve baseline performance monitor
run-ci
#13614
opened Nov 20, 2025 by
mickqian
Loading…
6 tasks
[Fix] Fix uvloop get_event_loop() is not suitable for 0.22.x
dependencies
Pull requests that update a dependency file
#13612
opened Nov 20, 2025 by
tom-jerr
Loading…
2 of 6 tasks
Test reorganization: Move tests to per_commit/ and manual/, add registration system
amd
deepseek
documentation
Improvements or additions to documentation
hicache
Hierarchical Caching for SGLang
lora
Multi-modal
multi-modal language model
npu
quant
LLM Quantization
run-ci
speculative-decoding
#13610
opened Nov 20, 2025 by
alisonshao
Loading…
[WIP] [Docs] Add docs for EP (expert parallelism)
documentation
Improvements or additions to documentation
#13609
opened Nov 20, 2025 by
JustinTong0323
Loading…
11 tasks
enable flashinfer fp8 gemm if deepgemm disabled
run-ci
#13608
opened Nov 19, 2025 by
b8zhong
Loading…
tiny remove deprecated endpoint call
documentation
Improvements or additions to documentation
run-ci
#13607
opened Nov 19, 2025 by
b8zhong
Loading…
Fix forward_metadata in DeepseekV2AttentionMLA to enable PD
deepseek
#13605
opened Nov 19, 2025 by
michael-amd
•
Draft
6 tasks
[AMD] Add 8-GPU MX35X test running DSR1-MXFP4 model for AMD CI
amd
deepseek
run-ci
#13602
opened Nov 19, 2025 by
hubertlu-tw
Loading…
6 tasks
[1/2] Refactor DeepGeem requant for FP8 Linear on Blackwell
deepseek
#13601
opened Nov 19, 2025 by
Fridge003
Loading…
6 tasks
enable csgmv automatically on cuda
documentation
Improvements or additions to documentation
lora
#13600
opened Nov 19, 2025 by
b8zhong
Loading…
test: Replace hardcoded CUDA device with get_device() for XPU support
#13599
opened Nov 19, 2025 by
kalyank007
Loading…
[router] Use qualified tool names (server_label__tool_name) in MCP inventory
model-gateway
run-ci
#13598
opened Nov 19, 2025 by
key4ng
Loading…
6 tasks
[kimi k2 thinking] Avoid useless torch.zeros_
sgl-kernel
#13596
opened Nov 19, 2025 by
BBuf
Loading…
2 of 8 tasks
feat: support execute customized tasks in model workers for second-level weight synchronization in RL systems
#13595
opened Nov 19, 2025 by
chaokunyang
Loading…
6 tasks
[Fix] fix KeyError when loading weights for Qwen3-Next-80B-A3B-Instruct-AWQ-4bit
#13594
opened Nov 19, 2025 by
Toseic
Loading…
6 tasks
fix: use server_args instead of ServerArgs to fetch base_gpu_id
#13593
opened Nov 19, 2025 by
leavelet
Loading…
1 of 6 tasks
[FEAT] optimize tensor zmq transfer for multimodal inputs
#13592
opened Nov 19, 2025 by
liusy58
Loading…
1 of 6 tasks
[Bugfix] Fix hidden state size in EAGLE PD disaggregation buffers
#13590
opened Nov 19, 2025 by
michelemarzollo
Loading…
2 of 6 tasks
[opt kimi k2 4 / n] Delete useless pad kernel in sgl_moe_align_block_size
#13587
opened Nov 19, 2025 by
BBuf
Loading…
6 tasks
[Ascend][Doc] updated installation guide for Ascend NPU
documentation
Improvements or additions to documentation
Enable RadixCache for Mamba2 models
run-ci
#13584
opened Nov 19, 2025 by
roikoren755
Loading…
2 of 6 tasks
[Fix] Remove multimodal_gen redundant get_bool_env_var func
#13583
opened Nov 19, 2025 by
shauntajoesph-ops
Loading…
6 tasks done
use class path string as custom logit processor parameter
deepseek
documentation
Improvements or additions to documentation
#13582
opened Nov 19, 2025 by
pengye91
Loading…
5 of 6 tasks
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.