Qwen2.5-Omni-SEA-R1

🌐 support textual cross-lingual processing, 🔊 Southeast Asia audio understanding, and 🧠 cross-modality reasoning.
📧 zhangly@i2r.a-star.edu.sg

We release Qwen2.5-Omni-SEA-R1, transferring the audio understanding to Southeast Asia languages of Singapore English, Singapore Chinese, Malay, and Indonesian, enhancing the multi-modality model with reasoning abilities. To our knowledge, this is the first multi-modality reasoning model targetting on SEA speech, and the first multi-modal model that performing deepseek-style reasoning on text, visual, and audio simutaneously.

We tested the model performance regarding the two improvements: ASR and Omni-reasoning.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support