Alibaba: qwen3-coder-plus API
The commercial flagship Mixture-of-Experts (MoE) coding model released by Alibaba Cloud on July 22, 2025. It serves as the enhanced version of the open-source Qwen3-Coder-480B-A35B-Instruct. With 480 billion total parameters and 35 billion activated (35B) per token, it is designed as the "Most Agentic Coder" to date. Optimized through extensive reinforcement learning for Agentic Coding, Browser-Use, and multi-tool orchestration, it can autonomously run tests, fix bugs, manage environments, and perform cross-file refactoring. Its performance rivals or exceeds top-tier proprietary models like Claude 4 and GPT-4o in complex software engineering workflows.
- Context window: 256,144 tokens
- Max output: 65,536 tokens
- Input: text
- Output: text
- Tool calling: Supported
- File input: Supported
- Released: 2025-09-22
- Knowledge cutoff: 2025-05-31
Frequently Asked Questions
What is the context window of qwen3-coder-plus?
qwen3-coder-plus supports a context window of up to 256,144 tokens.
Does qwen3-coder-plus support function calling?
Yes. qwen3-coder-plus supports tool / function calling.
What is the knowledge cutoff of qwen3-coder-plus?
The knowledge cutoff of qwen3-coder-plus is 2025-05-31.