Audex: Open-Source Audio-Text LLM with Leading Multimodal Capabilities
_weiping · x · 2026-07-08
The open-source unified audio-text LLM Audex has been released. It employs a single 30B-A3B MoE architecture to project audio inputs into the text embedding space. It achieves state-of-the-art performance among open-source models in audio understanding, speech recognition, translation, speech synthesis, audio generation, and speech-to-speech tasks. Furthermore, it outperforms the text-only Qwen3.5-35B-A3B in math and code reasoning, alignment, and instruction following.
Related event: NVIDIA Releases Open-Source Audio-Text LLM Audex-30B(9 posts)→
More from Multimodal
- Reddit Users Say the Ming Image Model Is Seriously Underrated — alley103 · 2026-10-27
- Riverside launches 4 AI agents to replace your entire video production crew — SucceededMind · 2026-10-27
- Creator ships absurd AI-edited videos on the daily — yungcontent · 2026-10-27
- An All-AI Music Video: Seedance for Video, Suno for Music, GPT6 Astra for Lyrics — qwer4790 · 2026-10-27
- MiniMax H3 video demo crawls through the Backrooms with near-film quality — warzone_afro · 2026-10-27
- AI-generated sitcom scene puts a sloth in a retro office interview — heresalexandria · 2026-10-27