v0.24.1

knavesec/CredMasterv0.24.1Jul 10, 2026by Gaohan123

AI Summary

This patch release addresses a performance regression in the Qwen-Image model and restores the default operator priority for RMSNorm to ensure correct model execution.

Key Highlights

  • Resolves performance regression for the Qwen-Image model
  • Fixes the default operator priority for RMSNorm
  • Restores vllm_c IR op priority and torch.nn.RMSNorm support

New Features

  • Restored vllm_c IR op priority
  • Restored torch.nn.RMSNorm support
  • Performance optimization for Qwen-Image model

Full Release Notes

## Highlights

This release resolves performance regression of Qwen-Image model according to issue #4964 . And it also fixes the default operator priority for RMSNorm.

## What's Changed
* [BugFix] Restore vllm_c IR op priority and torch.nn.RMSNorm for Qwen-Image (Same As #5009) by @NumberWan in https://github.com/vllm-project/vllm-omni/pull/5017


**Full Changelog**: https://github.com/vllm-project/vllm-omni/compare/v0.24.0...v0.24.1