Qwen: Qwen3 30B A3B

Qwen3-30B-A3B is a powerful Mixture-of-Experts language model from the Qwen series, featuring a smart switch between “thinking” and “dialogue” modes. With 30.5 billion total parameters (3.3 billion active), it excels at complex reasoning, multilingual processing, creative writing, mathematics, and coding. It uses YaRN to support context lengths up to 131,000 tokens and significantly outperforms earlier models in capability and versatility.