Run Qwen3-Coder-30B-A3B-Instruct and 480B-A35B locally with Unsloth Dynamic quants.
Qwen3-Coder is Qwen’s new series of coding agent models, available in 30B (Qwen3-Coder-Flash) and 480B parameters. Qwen3-480B-A35B-Instruct achieves SOTA coding performance rivalling Claude Sonnet-4, GPT-4.1, and Kimi K2, with 61.8% on Aider Polygot and support for 256K (extendable to 1M) token context.
We also uploaded Qwen3-Coder with native 1M context length extended by YaRN and full-precision 8bit and 16bit versions. Unsloth also now supports fine-tuning and RL of Qwen3-Coder.