AI Hub
All models
LLMOpen weights

Qwen3-Coder

Alibaba

Updated May 22, 2026 · Benchmarks via Artificial Analysis, specs & pricing via OpenRouter · Methodology · Report an error

Specifications

Type
LLM
Access
Open weights
Released
July 22, 2025
License
Apache 2.0
Parameters
480B (35B active)
Context window
262,144 tokens
Input
Text
Output
Text

Benchmarks

Our take

Qwen’s flagship open agentic-coding MoE, reaching near-parity with leading proprietary coders on SWE-bench; extendable to ~1M context via YaRN.

Related research

Links

Compare Qwen3-Coder with

See all Qwen3-Coder alternatives →