NVIDIA open-sources Nemotron-Labs-Diffusion-Uni, unified diffusion LM for multimodal
siyan_zhao · x · 2026-10-29
NVIDIA's Nemotron team released Nemotron-Labs-Diffusion-Uni, a unified diffusion language model handling both multimodal understanding and generation. It achieves SoTA on both via on-policy distillation from autoregressive VLMs and continuous flow models. Weights are open.
More from Multimodal
- AssemblyAI team goes live to discuss its newly shipped TTS-1 text-to-speech model — AssemblyAI · 2026-10-29
- Video2Game: H3 Max Generates Gameplay Video, Claude Turns It Into a Playable three.js Game — OdinLovis · 2026-10-29
- Xmax ships X2.1 real-time video editing at 1080P for $0.0016/sec, live face and outfit swaps — Scobleizer · 2026-10-29
- Genie 4 demo: chilling in a Japanese onsen with pettable capybaras — venturetwins · 2026-10-29
- Mirage launches Tesseract: turn any video into an agent-editable project — gizakdag · 2026-10-29
- PixVerse-Generated Cat Working Out on Playground Equipment Is Uncannily Convincing — Additional-Engine402 · 2026-10-29