Qwen3 Coder 30b A3B Instruct
Qwen3 Coder 30B A3B is a Mixture-of-Experts coding specialist with around 30.5 billion total parameters, activating only 3.3 billion per inference for efficient processing. It provides a 160K context window, extendable further with Yarn for large-codebase analysis. The model is tuned for code generation with tool use and structured output, supporting agentic coding workflows and OpenAI-compatible tool-call formats. Optimized for repository-scale code understanding and multi-file refactoring tasks, it is open-weight under the Apache 2.0 license, suited to local development and agentic code completion.
Key info
Benchmarks
Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.
Available routes
No routes currently available — Qwen3 Coder 30b A3B Instruct isn't routed through the Opper gateway right now. It may return.
Contact us about this model →