Qwen3 30b A3b Thinking moves to version Qwen3-30B-A3B-Thinking-2507

The OnAirToday AI Progress Tracker recorded a verified change to Alibaba's Qwen3 30b A3b Thinking (version Qwen3-30B-A3B-Thinking-2507): Version now Qwen3-30B-A3B-Thinking-2507 (was qwen3-30b-a3b-thinking).

Alibaba's Qwen3-30B-A3B-Thinking-2507 is an open-weight, text-only reasoning model with 30.5 billion total parameters and 3.3 billion activated per token. This thinking-only edition provides a native 262,144-token context and is distinct from the original hybrid Qwen3-30B-A3B and the separate Instruct-2507 edition.

What changed in this review:

• Identified the precise Thinking-2507 edition, distinct from the original hybrid model and the separate Instruct-2507 edition.
• Added the documented native context window of 262,144 tokens.
• Updated GPQA Diamond to the vendor-published 73.4% and added the vendor-published AIME 2025 result of 85.0%.
• Capability score 70 → 79
• Flaws and limitations updated
• Risks and threats updated

Teams building on Qwen3 30b A3b Thinking 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/qwen3-30b-a3b-thinking.

References:

• https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507
• https://huggingface.co/Qwen/Qwen3-30B-A3B