# vLLM adds Mistral Large 4 support with reasoning effort levels

> Original title: The release adds low / medium / high reasoning effort, defaulting to high. Pixtral vision supports multiple images per prompt for multimo...

- Source: vLLM
- Published: 2026-10-29T15:13:43.000Z
- Bad Influence: https://badinfluence.news/items/rp05sr2x5jnioe2dl63ck0ade
- Original: https://x.com/vllm_project/status/2115824409219105019

## Summary

vLLM's release adds low, medium, and high reasoning effort settings, with high as the default. Pixtral vision now accepts multiple images per prompt. SkyRL also supports day-0 reinforcement learning training for Mistral Large 4 using vLLM rollouts.

## Post

The release adds low / medium / high reasoning effort, defaulting to high. Pixtral vision supports multiple images per prompt for multimodal workflows.

SkyRL (@NovaSkyAI) also supports day-0 RL training for Mistral Large 4, with vLLM rollouts.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHVzrFqjasAAIj_N.jpg&mode=thumb&exp=1793491200&sig=52180d29423775ba)