vLLM v0.25.1

v0.25.1
Fixed 2
  • Avoid blocking model launching when no system FFmpeg is available for TorchCodec by deferring the error to runtime instead of raising it at import time
  • Guard mixed-dtype allreduce RMSNorm quant fusions to prevent hidden state corruption when activation and RMSNorm weight dtypes differ

vLLM v0.25.1

Highlights

This release features 2 commits from 2 contributors (1 new)!

v0.25.1 is a patch release containing two targeted bug fixes on top of v0.25.0.

Bug Fixes
  • Avoid blocking model launching when no system FFmpeg is available for TorchCodec (#47888). Previously import torchcodec raised a RuntimeError at import time when system FFmpeg was missing, which blocked startup (e.g. vllm serve Qwen/Qwen3-VL-2B-Instruct) even when TorchCodec was not in use. The error is now deferred to runtime so it only surfaces if TorchCodec is actually needed.
  • Guard mixed-dtype allreduce RMSNorm quant fusions (#48330). The fused FlashInfer allreduce + RMSNorm + static-quantization patterns could match graphs where the activation and RMSNorm weight dtypes differ (e.g. a BF16 residual stream with an FP32 Gemma/Qwen-style RMSNorm weight in NVFP4 models), corrupting the hidden state and producing garbage output such as repeated !!!!! tokens. A dtype-match guard now routes incompatible mixed-dtype graphs to the safe path, while same-dtype models retain the full allreduce + RMSNorm + quant fusion.
Contributors

@Isotr0py, @hugo-cen

New Contributors
View original

Upgraded? How did it go?

Discussion