Qwen2-Audio

Parameters

8.2B

Published

Jul 15, 2024

Qwen2-Audio is an AI model developed by Alibaba (China), first published in July 2024. It works in the multimodal, language and audio domain, on tasks such as speech recognition (asr), speech synthesis, translation and speech-to-speech.

Epoch AI has no training-compute estimate for this model. The model has 8,200,000,000 parameters.

Access: Open weights (unrestricted). Its weights are openly available. It is built on top of Qwen-7B. Epoch AI rates the confidence of this record as likely.

Full record

Organization: Alibaba
Country of organization: China
Domain: Multimodal, Language, Audio
Task: Speech recognition (ASR), Speech synthesis, Translation, Speech-to-speech
Compute estimation method: Operation counting
Parameters: 8,200,000,000
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Base model: Qwen-7B
Epoch confidence: Likely