Qwen Team Launches Qwen3-Coder for Advanced Agentic Coding Tasks

Qwen Team has announced Qwen3-Coder, a new family of agentic code models designed for long-context, multi-step programming tasks. The flagship variant, Qwen3-Coder-480B-A35B-Instruct, is a Mixture-of-Experts (MoE) model with 480 billion total parameters and 35 billion active parameters per forward pass. It supports 256K tokens natively and up to 1 million tokens via context extension, targeting repository-scale inputs and extended tool interactions.

Key Features

• Agentic Focus: Unlike static code generation models, Qwen3-Coder emphasizes execution and decision-making, trained using reinforcement learning on real-world tasks where success is defined by code execution and problem-solving.
• Scaled Training: The team deployed a system capable of running 20,000 parallel environments on cloud infrastructure to simulate developer workflows.
• Open Tooling: Qwen released Qwen Code, an open-source CLI forked from Gemini CLI, with enhanced tool use and function calling support. It’s installable via npm and compatible with OpenAI APIs.

Availability

Qwen3-Coder is currently accessible via DashScope API, with international endpoints for developers outside mainland China. Sample Python code is provided for quick integration. Smaller model variants are expected soon to reduce inference costs.

"Qwen3-Coder’s local use isn’t a cost-saver unless you’ve got the right multi-GPU setup. Running smaller versions when they release might lower expenses." — Reddit user feedback

Future Work

The team plans to expand the Qwen Coding Agent’s capabilities and explore self-improvement mechanisms for iterative performance upgrades with minimal human supervision.

About the Author

Robert Krzaczyński