Qwen3 235B A22B: separate Instruct-2507 and Thinking-2507 successors shipped in July 2025; this profile remains about the original hybrid-thinking release
The original hybrid-thinking release, distinct from its July 2025 successors
Separate Instruct-2507 and Thinking-2507 successors shipped in July 2025; this profile remains about the original hybrid-thinking release — 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): Separate Instruct-2507 and Thinking-2507 successors shipped in July 2025; this profile remains about the original hybrid-thinking release.
Alibaba's Qwen3-235B-A22B is an open-weight, text-only mixture-of-experts model with 235 billion total parameters and 22 billion activated per token, supporting both thinking and non-thinking responses. Separate Instruct-2507 and Thinking-2507 successors were released in July 2025; this profile covers the original hybrid-thinking model.
What changed in this review:
- Benchmark coverage now uses Alibaba's published thinking-mode results: GPQA Diamond, AIME 2024, AIME 2025 and LiveCodeBench v5.
- Removed MATH Level 5, Aider Polyglot and LMArena Text entries without established model-specific result citations.
- Flaws and limitations updated
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
The distinction keeps the original hybrid-thinking release separate from its later successors. Clear versioning matters when readers compare Qwen3 model profiles and documentation.

Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.