Qwen: Qwen3 30B A3B Thinking 2507
Analysis Summary
Qwen3 30B A3B Thinking 2507 is the extended-thinking variant of Alibaba's Qwen3 30B A3B MoE model, with an intelligence index of 15.8 and a coding index of 14.6. Its instruction-following score of 0.507 and long-context reasoning score of 0.59 are stronger than the instruct variant, suggesting the thinking mode improves structured output reliability. Tool use and function calling are supported.
For business use, this model is best suited to tasks where deliberate reasoning improves output quality, such as structured analysis, summarisation, or content generation over longer documents. Its agentic index of 16.7 is low, making it a poor fit for multi-step agent workflows. Coding performance is also limited.
At $0.08 input and $0.40 output, pricing is reasonable for a thinking-mode model. A -4 point regional penalty applies. Teams needing a cheap reasoning-mode model for structured content tasks may find value here, but stronger alternatives exist at similar price points for most business workloads.
Assessed June 17, 2026
Editorial notes
Qwen3 30B A3B Thinking 2507 is a reasoning-mode variant with improved instruction following and long-context scores, but low agentic and coding indices limit its business utility.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Qwen: Qwen3 30B A3B Thinking 2507 compares
Qwen: Qwen3 30B A3B Thinking 2507 ranks #163 of 380 AI models we track for overall intelligence, #185 of 317 for coding, #203 of 292 for agentic tasks. Its 131K-token context window is larger than 59% of the models we list. At $0.08 per million input tokens it is cheaper than 68% of comparable models.
About Qwen: Qwen3 30B A3B Thinking 2507
Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for āthinking mode,ā where internal reasoning traces are separated..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Qwen: Qwen3 30B A3B Thinking 2507 stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
qwen/qwen3-30b-a3b-thinking-2507
|
| Provider | qwen |
| Release Date | August 28, 2025 |
| Context Length | 131,072 tokens |
| Max Completion | 131,072 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.08 | $0.000080 |
| Output | $0.40 | $0.000400 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Qwen: Qwen3 30B A3B Thinking 2507
How much does Qwen: Qwen3 30B A3B Thinking 2507 cost?
Qwen: Qwen3 30B A3B Thinking 2507 costs $0.08 per million input tokens and $0.40 per million output tokens.
What is the context window of Qwen: Qwen3 30B A3B Thinking 2507?
Qwen: Qwen3 30B A3B Thinking 2507 has a context window of 131,072 tokens (131K).
Is Qwen: Qwen3 30B A3B Thinking 2507 good for coding?
On our coding benchmark index, Qwen: Qwen3 30B A3B Thinking 2507 ranks #185 of 317 models, placing it in the broader range of the field for code generation and debugging.
What can Qwen: Qwen3 30B A3B Thinking 2507 do?
Qwen: Qwen3 30B A3B Thinking 2507 supports tool use and function calling.
Who created Qwen: Qwen3 30B A3B Thinking 2507?
Qwen: Qwen3 30B A3B Thinking 2507 is developed by Qwen and was released on August 28, 2025.
