Skip to main content

Qwen3-Coder-30B-Instruct

Description​

“Qwen3-Coder-30B-Instruct” is a 30 billion-parameter Mixture-of-Experts (MoE) model engineered for autonomous agentic coding and seamless tool integration. It dynamically routes tasks to specialized expert subnetworks—covering syntax generation, error diagnostics, and API handling—to maximize efficiency and context awareness. Instruction-tuned on real-world development workflows, it excels at orchestrating complex, multi-step programming and automation pipelines.

The following limitations apply:

  • Maximum context length: 65,536 tokens
  • This model does not support thinking / reasoning

It is recommended to operate the model with the following parameters:

  • temperature=0.7
  • top_p=0.8
  • top_k=20
  • repetition_penalty=1.05

Terms of use and licensing​

The general terms of use apply. The model is provided by Alibaba under the Apache 2.0 License, and reuse of the generated content is not subject to any additional restrictions.