Qwen: Qwen3 Coder 30B A3B Instruct
Analysis Summary
Qwen3 Coder 30B A3B Instruct is a sparse MoE model from Alibaba's Qwen team, optimised for code generation with a 160K context window. Its coding index of 19.4 and livecodebench score of 0.403 are below mid-tier, and its intelligence index of 13.6 limits its usefulness for complex reasoning tasks. Tool use and function calling are supported, but the agentic index of 24.8 is low.
For business use, this model is best suited to simple code completion, boilerplate generation, or cost-sensitive pipelines where a small, cheap model is preferable to a larger one. Its math and general reasoning scores are weak, so it should not be used for analytical or multi-step agentic tasks. Instruction following is also below average.
At $0.07 input and $0.27 output, pricing is reasonable for a coding-focused model, but stronger alternatives exist at similar price points. A -4 point regional penalty applies given limited enterprise availability. Teams with very specific, narrow coding automation needs may find it useful, but most businesses will benefit from a more capable model.
Assessed June 17, 2026
Editorial notes
Qwen3 Coder 30B A3B is a coding-focused MoE model with tool use support, but limited reasoning depth and low agentic reliability make it suitable only for narrow, cost-sensitive coding tasks.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Qwen: Qwen3 Coder 30B A3B Instruct compares
Qwen: Qwen3 Coder 30B A3B Instruct ranks #180 of 380 AI models we track for overall intelligence, #152 of 317 for coding, #162 of 292 for agentic tasks. Its 160K-token context window is larger than 59% of the models we list. At $0.07 per million input tokens it is cheaper than 70% of comparable models.
About Qwen: Qwen3 Coder 30B A3B Instruct
Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use. Built on the..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Qwen: Qwen3 Coder 30B A3B Instruct stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
qwen/qwen3-coder-30b-a3b-instruct
|
| Provider | qwen |
| Release Date | July 31, 2025 |
| Context Length | 160,000 tokens |
| Max Completion | 32,768 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.07 | $0.000070 |
| Output | $0.27 | $0.000270 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Qwen: Qwen3 Coder 30B A3B Instruct
How much does Qwen: Qwen3 Coder 30B A3B Instruct cost?
Qwen: Qwen3 Coder 30B A3B Instruct costs $0.07 per million input tokens and $0.27 per million output tokens.
What is the context window of Qwen: Qwen3 Coder 30B A3B Instruct?
Qwen: Qwen3 Coder 30B A3B Instruct has a context window of 160,000 tokens (160K).
Is Qwen: Qwen3 Coder 30B A3B Instruct good for coding?
On our coding benchmark index, Qwen: Qwen3 Coder 30B A3B Instruct ranks #152 of 317 models, placing it in the broader range of the field for code generation and debugging.
What can Qwen: Qwen3 Coder 30B A3B Instruct do?
Qwen: Qwen3 Coder 30B A3B Instruct supports tool use and function calling.
Who created Qwen: Qwen3 Coder 30B A3B Instruct?
Qwen: Qwen3 Coder 30B A3B Instruct is developed by Qwen and was released on July 31, 2025.
