Novita AI
Qwen-MT-Plus is the Qwen series' flagship translation model, now fully upgraded to the Qwen3 architecture and supporting 92 languages for highly accurate, natural-sounding translations. The Alibaba Cloud Model Studio documentation confirms strong contextual understanding, terminology control, and format preservation suited to specialized domains. The model is text-in and text-out only, with an 8,192-token max input, 8,192-token max output, and a 16,384-token context window. Function calling, structured outputs, web search, prefix completion, context caching, batch inference, and fine-tuning are all unsupported, while rate limits sit at 60 RPM with 25,000 or 100,000 TPM depending on region.
