Baizhou Zhang
Baizhou Zhang
@YAMY1234 Thanks~ Since this PR will break the usage of deepseek v32, can you please change all the related usage (appending --dp argument) in the test cases (files with prefix...
/tag-and-rerun-ci
> [@b8zhong](https://github.com/b8zhong) I saw you self-assigned, am I still good to work on this? @b8zhong will work on this. Thanks anyway
Setting SGLANG_MOE_NVFP4_DISPATCH=1 for prefill node should solve this. There has been some refactors on MoE recently. https://github.com/sgl-project/sglang/pull/13715#issuecomment-3566029238
All dpsk v3.2 related tests passed https://github.com/sgl-project/sglang/actions/runs/19608731556?pr=13718
/tag-and-rerun-ci
seems flashinfer-cubin v0.5.3 is not ready: https://github.com/flashinfer-ai/flashinfer/issues/2133 temporarily remove it from pyproject.toml
@FlamingoPg Seems triton is failing with this error https://github.com/sgl-project/sglang/actions/runs/19317390340/job/55254572458?pr=12969 We have met similar situations before. It's caused by https://github.com/triton-lang/triton/pull/8536/files, and we solved it by pinning the version of Triton to...
Updates related to Torch 2.9 - Upgrade dependencies in pyproject.toml/Dockerfiles. Torch -> 2.9.1, triton -> 3.5.1 - Skip flaky multimodal test cases: `Qwen/Qwen-Image-Edit`, `Wan-AI/Wan2.2-I2V-A14B-Diffusers` - Forcefully reinstall nvidia-cudnn-cu12==9.16.0.29, to avoid...