Alibaba has released Qwen2-Math for solving mathematical problems and Qwen2-Audio for multi-modal voice or text input processing.
Qwen2-Math is available in base and instruction-tuned versions with 1.5 billion, 7 billion, or 72 billion parameters. Qwen2-Audio can operate in Voice Chat or Audio Analysis modes, supporting text and audio input but only outputting text.
Both model families are based on Alibaba's Qwen2 LLM series and are available under the Apache 2.0 license, except for the largest version of Qwen2-Math.
The company plans to release bilingual models supporting English and Chinese, with multilingual models in development.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.