sgl-project/sglang v0.5.15.post1
v0.5.15.post1 includes a few patches, mostly for GLM 5.2
Key points
- #30454 #30627: Fix DSA model launching on non Cuda/HIP devices
- #30858: Fix flashinfer dependency on Cuda 12 images
- #31001: Fix NaN outputs caused by flashinfer trtllm FP4 MoE kernels on long input
- #30992: Fix GLM 5.2 IndexShare on Context Parallel setting
Sources (1)
- [1]sgl-project/sglang v0.5.15.post1GitHub: sgl-project/sglang · Jul 14, 08:43 AM
v0.5.15.post1 includes a few patches, mostly for GLM 5.2
- #30454 #30627: Fix DSA model launching on non Cuda/HIP devices
Extractive summary: sentences quoted from the sources.