Qwen3 235B A22B moves to version Qwen3-235B-A22B
Tracker verifies the model’s updated name against Alibaba documentation
Version now Qwen3-235B-A22B (was qwen3-235b-a22b) — verified by the OnAirToday AI Progress Tracker against Alibaba's published documentation.
OnAirToday
Newsroom
The OnAirToday AI Progress Tracker recorded a verified change to Alibaba's Qwen3 235B A22B (version Qwen3-235B-A22B): Version now Qwen3-235B-A22B (was qwen3-235b-a22b).
Alibaba's Qwen3-235B-A22B is an Apache-2.0-licensed, open-weight text model with 235 billion total parameters and 22 billion activated per token, combining thinking and non-thinking modes. Separate Instruct-2507 and Thinking-2507 successors shipped in July 2025; this profile describes the original hybrid-thinking release.
What changed in this review:
- Separate Instruct-2507 and Thinking-2507 successors shipped in July 2025; this profile remains about the original hybrid-thinking release.
- Added Alibaba's published AIME 2025 result of 81.5%.
Teams building on Qwen3 235B A22B should confirm limits, pricing and benchmark coverage against the vendor's own documentation before relying on these figures.
The full profile — capabilities, benchmark citations, known flaws and risk guidance — is on the tracker at /ai-models/qwen3-235b-a22b.
References:
- https://huggingface.co/Qwen/Qwen3-235B-A22B
- https://qwenlm.github.io/blog/qwen3/
- https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507
- https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507
Generated insight by OnAirToday AI
Consistent model naming helps researchers distinguish documented versions from earlier labels. The tracker’s verification against Alibaba documentation provides a reference for the updated name.
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.