- Notifications
You must be signed in to change notification settings - Fork4.5k
Pull requests: sgl-project/sglang
Author
Uh oh!
There was an error while loading.Please reload this page.
Label
Uh oh!
There was an error while loading.Please reload this page.
Projects
Uh oh!
There was an error while loading.Please reload this page.
Milestones
Uh oh!
There was an error while loading.Please reload this page.
Reviews
Assignee
Assigned to nobodyLoading
Uh oh!
There was an error while loading.Please reload this page.
Sort
Pull requests list
Skip Dynamo tracing in symmetric-memory NCCL paths
#19080 openedFeb 20, 2026 bymmangkadLoading…
5 tasks
Document & auto-enable FP8 block-wise CUTLASS GEMM for SM120 documentationImprovements or additions to documentation
#19078 openedFeb 20, 2026 byArush04Loading…
2 of 5 tasks
[Refactor] Benchmark Phase 1: extract utils and datasets from bench_serving
#19077 openedFeb 20, 2026 byRatish1Loading…
2 of 5 tasks
[Fix] Quick fix for int32 overflow in Mooncakes' send_kvcache_slice run-ci
#19076 openedFeb 20, 2026 byYAMY1234Loading…
5 tasks
[Diffusion][BUG] Fix reading multiple prompts from prompt file diffusionSGLang Diffusion
#19075 openedFeb 20, 2026 bysushildubey171Loading…
3 of 5 tasks
fix: disable structural_tag for function_call_parser.GptOssDetector to fix 500error when using tools + streaming
#19074 openedFeb 20, 2026 bysokolgoodLoading…
2 of 5 tasks
fix(dense): fix Qwen3.5 dense model precision bug in TP_SIZE>1 run-ci
#19070 openedFeb 20, 2026 byzju-stu-lizhengLoading…
Add missing prefill batch log in disaggregation prefill mode
#19067 openedFeb 20, 2026 byKangyan-ZhouLoading…
2 tasks
Optimization of Qwen Image, Qwen 2.5 ViT and LLM diffusionSGLang Diffusion Multi-modalmulti-modal language model
#19066 openedFeb 20, 2026 byzhyajieLoading…
3 tasks done
Refactor dumper and change on_forward_pass_start API
#19065 openedFeb 20, 2026 byfzyzcjyLoading…
5 tasks
[Diffusion] Restruct and clean Diffusion rotary embedding diffusionSGLang Diffusion run-ci
#19064 openedFeb 20, 2026 byBBufLoading…
5 tasks
[jit_kernel] Add JIT tree_speculative_sampling_target_only kernel speculative-decoding
#19061 openedFeb 20, 2026 byJohnsonmsLoading…
[diffusion] refactor: reduce redundancy and improve stage api diffusionSGLang Diffusion documentationImprovements or additions to documentation lora npu run-ci
#19060 openedFeb 20, 2026 bymickqianLoading…
5 tasks
[jit_kernel] Add fused_qknorm_rope JIT kernel run-ci
#19059 openedFeb 20, 2026 byJohnsonmsLoading…
4 of 5 tasks
[jit_kernel] Add prepare_moe_input JIT kernels (port from sgl-kernel AOT) npu
#19058 openedFeb 20, 2026 byJohnsonmsLoading…
[jit_kernel] Add moe_sum JIT kernel (port from sgl-kernel AOT) run-ci
#19057 openedFeb 20, 2026 byJohnsonmsLoading…
Migrate moe_sum_reduce from sgl-kernel AOT to JIT
#19056 openedFeb 20, 2026 byJohnsonmsLoading…
5 tasks
ProTip! Addingno:label will show everything without a label.