release aidev NEEDS REVIEW SIG 1/5

Xiaomi MiMo V2.5 ASR speech recognition model published

Xiaomi published an automatic-speech-recognition variant of MiMo V2.5 supporting Chinese, English, and Cantonese. Licensed under MIT with endpoint compatibility.

PUBLISHED2026-04-23
OBSERVED2026-08-11
AGE4mo
SOURCES1
  • Pipeline: automatic-speech-recognition via Transformers
  • Languages: zh, en, yue (Cantonese)
  • License: MIT
  • Format: safetensors
  • Tags: endpoints_compatible, text-generation-inference
  • Hosted on Hugging Face under XiaomiMiMo organisation

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.