MOSS Audio Transcription Model Integrates with vLLM

vllm_project · x · 2026-07-09

The vLLM team congratulated MosiAI on releasing MOSS-Transcribe-Diarize-0.9B, an open-source, end-to-end long audio transcription model that supports multi-speaker recognition and timestamp output, achieving day-one support in vLLM.

Related event: MOSS Releases Open-Source 0.9B Audio Transcription Model Supported by vLLM(5 posts)→

Original post →

More from Infra

Infra channel →