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
- 3D ResNet Paper Crosses 3,000 Citations Eight Years After CVPR 2018 — HirokatuKataoka · 2026-09-11
- Non-coder builds full-featured Android ComfyUI client with ChatGPT, submits to Google Play — ComfierUI · 2026-09-11
- RunningHub open-sources H3Lightning, speeding up MiniMax H3 video generation 12x — 智东西 · 2026-09-11
- FastH3-Live hits 22fps: acceleration node benchmarks and the --vram-headroom trick — spartong945 · 2026-09-11
- Midjourney style code share: --sref 2912175708 — tisch_eins · 2026-09-11
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11