Qwen: Qwen3 Coder 30B A3B Instruct

Qwen3-Coder-30B-A3B-Instruct is a cutting-edge 30.5B parameter Mixture-of-Experts (MoE) model, featuring 128 experts with 8 active per forward pass. It is specifically engineered for advanced code generation, comprehensive repository-scale understanding, and sophisticated agentic tool use. Built upon the robust Qwen3 architecture, this model boasts a native context length of 256K tokens, extendable up to 1M with Yarn, ensuring it can handle vast amounts of information. This model excels in tasks requiring precise instruction-following without a 'thinking mode', and seamlessly integrates with OpenAI-compatible tool-use formats. Its capabilities include strong performance in function calls, browser use, and structured code completion. With a context window of 262K tokens and a max output of 4K tokens, it's ideal for complex coding challenges. Pricing is competitive at $0.07/0.27 per 1M tokens (input/output), and it's available for free access on Multi AI.