Qwen: Qwen3 30B A3B Instruct 2507
Analysis Summary
Qwen3 30B A3B Instruct 2507 is a sparse mixture-of-experts model from Alibaba's Qwen team, activating only 3B parameters per forward pass. It supports tool use and function calling with a 131K context window, but its intelligence and coding indices sit at the lower end of the benchmarked field, limiting its usefulness for complex reasoning or autonomous workflows.
For business use, this model is best suited to lightweight automation, simple classification, or cost-sensitive pipelines where raw reasoning depth is not critical. Its agentic score is very low, making it a poor fit for multi-step agent tasks or complex code generation. Instruction following is also below average.
At under $0.05 per million input tokens, it offers reasonable cost efficiency for narrow, well-defined tasks. Teams looking for a cheap MoE option for simple structured outputs may find value here, but most business workloads will benefit from a more capable model.
Assessed June 17, 2026
Editorial notes
Qwen3 30B A3B Instruct 2507 is a compact MoE model with tool use and function calling, but limited reasoning and agentic depth place it well below mid-tier for demanding business tasks.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Qwen: Qwen3 30B A3B Instruct 2507 compares
Qwen: Qwen3 30B A3B Instruct 2507 ranks #236 of 380 AI models we track for overall intelligence, #192 of 317 for coding, #271 of 292 for agentic tasks. Its 131K-token context window is larger than 59% of the models we list. At $0.05 per million input tokens it is cheaper than 73% of comparable models.
About Qwen: Qwen3 30B A3B Instruct 2507
Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and..
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 Instruct 2507 stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
qwen/qwen3-30b-a3b-instruct-2507
|
| Provider | qwen |
| Release Date | July 29, 2025 |
| Context Length | 131,072 tokens |
| Max Completion | 32,000 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.05 | $0.000048 |
| Output | $0.19 | $0.000193 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Qwen: Qwen3 30B A3B Instruct 2507
How much does Qwen: Qwen3 30B A3B Instruct 2507 cost?
Qwen: Qwen3 30B A3B Instruct 2507 costs $0.05 per million input tokens and $0.19 per million output tokens.
What is the context window of Qwen: Qwen3 30B A3B Instruct 2507?
Qwen: Qwen3 30B A3B Instruct 2507 has a context window of 131,072 tokens (131K).
Is Qwen: Qwen3 30B A3B Instruct 2507 good for coding?
On our coding benchmark index, Qwen: Qwen3 30B A3B Instruct 2507 ranks #192 of 317 models, placing it in the broader range of the field for code generation and debugging.
What can Qwen: Qwen3 30B A3B Instruct 2507 do?
Qwen: Qwen3 30B A3B Instruct 2507 supports tool use and function calling.
Who created Qwen: Qwen3 30B A3B Instruct 2507?
Qwen: Qwen3 30B A3B Instruct 2507 is developed by Qwen and was released on July 29, 2025.
