NVIDIA Releases Audex Audio Language Model

_weiping · x · 2026-07-08

The NVIDIA research team has officially released Audex, a unified audio-text large language model capable of understanding, generating, and reasoning across multiple modalities including text and speech. Achieving state-of-the-art performance on audio intelligence benchmarks, it represents a significant advancement in the multimodal AI field.

Related event: NVIDIA Releases Open-Source Audio-Text LLM Audex-30B(9 posts)→

Original post →

More from Multimodal

Multimodal channel →