Jin10 data reported on July 23, that in the early morning, Tongyi Qianwen officially released Qwen3-Coder, which is its most capable code model to date. Tongyi Qianwen states that Qwen3-Coder is available in multiple sizes, but we are eager to provide everyone with the current most powerful version, Qwen3-Coder-480B-A35B-Instruct. This is a 480B parameter model activating 35B parameters, natively supporting a 256K token context and can be expanded to 1M tokens via YaRN, with outstanding coding and agent capabilities. Qwen3-Coder-480B-A35B-Instruct has achieved SOTA results of open-source models in Agentic Coding, Agentic Browser-Use, and Agentic Tool-Use, and can compete with ClaudeSonnet4.
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
Tongyi Qianwen releases its most capable agent model Qwen3-Coder
Jin10 data reported on July 23, that in the early morning, Tongyi Qianwen officially released Qwen3-Coder, which is its most capable code model to date. Tongyi Qianwen states that Qwen3-Coder is available in multiple sizes, but we are eager to provide everyone with the current most powerful version, Qwen3-Coder-480B-A35B-Instruct. This is a 480B parameter model activating 35B parameters, natively supporting a 256K token context and can be expanded to 1M tokens via YaRN, with outstanding coding and agent capabilities. Qwen3-Coder-480B-A35B-Instruct has achieved SOTA results of open-source models in Agentic Coding, Agentic Browser-Use, and Agentic Tool-Use, and can compete with ClaudeSonnet4.