Qwen3 Coder 30B A3B Instruct

qwen/qwen3-coder-30b-a3b-instruct

Qwen3 Coder 30B A3B Instruct is a text‑only model from Qwen released on 2025‑07‑31. It supports tool usage, does not support reasoning, and provides an open‑weights implementation with a context length of 160 000 tokens. Pricing is $0.07 per million input tokens and $0.27 per million output tokens.

  • tool use
  • structured outputs
  • open-weight

At a glance

providerqwen
releasedJul 31, 2025
context262K
input / M$0.070

Specifications

What the API gives you

Context window
262K tokens
Max output
33K tokens
Input price
$0.070 / 1M tokens
Output price
$0.27 / 1M tokens
Cache read
Modalities
text → text
Knowledge cutoff
Jun 30, 2025
Released
Jul 31, 2025

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. This means the available task-specific evidence did not support a ranking yet—not that the model was reviewed and rejected.

Measured evidence

Public benchmark record

SWE-bench Verified
60.4%2025-09-01

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Self-reported scaffold plus model result. Treat it as a claim, not an independent measurement.

Open-weight model on Hugging Face.

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.