Qwen3 30B A3B Instruct 2507
Qwen · → text
Qwen3 30B A3B Instruct 2507 is Qwen's mixture-of-experts language model, designed for high-quality instruction following and multilingual understanding. It features 30.5B total parameters with 3.3B active parameters per inference and operates in non-thinking mode for text generation.
Input$0.1028 /M
Output$0.4112 /M