Qwen3 Technical Report

Team, 2025, Qwen3 Technical Report (arXiv preprint arXiv:2505.09388)

Subscriber deck — free during early access

1 cards in “4 Post-training” · whole deck

4 Post-training

What are the four post-training stages of Qwen3’s flagship models?
(1) Long chain-of-thought cold start; (2) reasoning RL; (3) thinking mode fusion: SFT that adds non-thinking behavior to the reasoning model; (4) general RL across many tasks. Smaller models skip these and are distilled from the large ones instead.
- Team, 2025, Qwen3 Technical Report (arXiv preprint arXiv:2505.09388), §4

Source

Qwen Team. 2025. Qwen3 Technical Report. In arXiv preprint arXiv:2505.09388. https://arxiv.org/abs/2505.09388