sglang

v0.5.15.post1 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 12d Model Serving & MLOps

✓ No known CVEs patched

✓ No known CVEs patched in this version

Topics

attention blackwell cuda deepseek diffusion glm

+12 more

gpt-oss inference llama llm minimax moe qwen qwen-image reinforcement-learning transformer vlm wan

Summary

AI summary

Fixed NaN outputs from flashinfer trtllm FP4 MoE kernels on long inputs.

Type	Severity	Summary	CVE
Bugfix
Bugfix	Medium	Fixes NaN outputs from flashinfer trtllm FP4 MoE kernels on long inputs. Fixes NaN outputs from flashinfer trtllm FP4 MoE kernels on long inputs. Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	Fixes GLM 5.2 IndexShare issues on PD disaggregation setting. Fixes GLM 5.2 IndexShare issues on PD disaggregation setting. Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	Fixes GLM 5.2 IndexShare issues on Context Parallel setting. Fixes GLM 5.2 IndexShare issues on Context Parallel setting. Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	Fixes DSA model launching on non Cuda/HIP devices. Fixes DSA model launching on non Cuda/HIP devices. Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	Fixes flashinfer dependency on Cuda 12 images. Fixes flashinfer dependency on Cuda 12 images. Source: llm_adapter@2026-07-14 Confidence: high	—

Full changelog

v0.5.15.post1 includes a few patches, mostly for GLM 5.2

#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
#30839: Fix GLM 5.2 IndexShare on PD disaggregation setting
#30992: Fix GLM 5.2 IndexShare on Context Parallel setting

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track sglang

Get notified when new releases ship.

About sglang

SGLang is a high-performance serving framework for large language models and multimodal models.