●Change8

Migrating to Unsloth September-2025-v3

Version September-2025-v3 introduces 1 breaking change. This guide details how to update your code.

Released: 9/26/2025

1
Breaking Changes
2
Migration Steps
2
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

Transformers inference code (rewritten for gpt-oss RL)BERT

Breaking Changes

●Issue #1

Transformers inference code was rewritten to enable faster RL inference for gpt-oss, as it is not yet vLLM compatible. Users relying on the previous inference implementation for gpt-oss RL might see changes in behavior or performance characteristics.

Migration Steps

  1. 1
    Ensure you have the latest version of transformers installed to support fine-tuning Qwen3 models.
  2. 2
    Do NOT use Flash Attention 3 when training gpt-oss models, as it will result in incorrect training loss.

Release Summary

This release introduces significant performance gains and new capabilities for gpt-oss Reinforcement Learning, alongside support for new models like DeepSeek-V3.1-Terminus and Magistral 1.2. Several bug fixes were implemented, including resolving issues with BERT and QAT + LoRA fast path.

Need More Details?

View the full release notes and all changes for Unsloth September-2025-v3.

View Full Changelog