Xiaomi MiMo publishes MiMo-Audio-7B-Instruct
Xiaomi MiMo released MiMo-Audio-7B-Instruct, the instruction-tuned version of their 7B parameter any-to-any multimodal model for audio and text. Published on Hugging Face under MIT license.
PUBLISHED2025-09-18
OBSERVED2026-08-11
AGE11mo
SOURCES1
- Instruction-tuned version of MiMo-Audio-7B-Base
- 7B parameter any-to-any multimodal model
- Modalities: Audio-to-Text, Text-to-Audio, Audio-to-Audio, Text-to-Text, Audio-Text-to-Text
- Based on Qwen2 architecture
- License: MIT
- Pipeline: any-to-any
COMMUNITY
No curated reactions recorded for this event. Facts and takes are kept in separate layers — community context is added by hand, never blended into the record above.