Qwen3 30B A3B: separate Instruct-2507 and Thinking-2507 editions shipped in July 2025; this profile remains about the original hybrid checkpoint

The OnAirToday AI Progress Tracker recorded a verified change to Alibaba's Qwen3 30B A3B (version Qwen3-30B-A3B): Separate Instruct-2507 and Thinking-2507 editions shipped in July 2025; this profile remains about the original hybrid checkpoint.

Qwen3-30B-A3B is Alibaba's open-weight, text-only mixture-of-experts model, with 30.5 billion total parameters, 3.3 billion activated parameters, and selectable thinking and non-thinking modes. Separate Instruct-2507 and Thinking-2507 editions were released in July 2025; this profile covers the original hybrid checkpoint.

What changed in this review:

• Benchmark entries now use vendor-published thinking-mode results for GPQA Diamond, AIME 2024, AIME 2025, and LiveCodeBench v5.
• API pricing: Not yet confirmed by public reporting.
• Capability score 62 → 68

Teams building on Qwen3 30B A3B 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.

References:

• https://huggingface.co/Qwen/Qwen3-30B-A3B
• https://arxiv.org/abs/2505.09388
• https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507
• https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507