Intern Releases Lumina U2: Diffusion LLM Unifying Video, 3D Understanding and Image Generation
bdsqlsz · x · 2026-09-05
Intern released Lumina U2, a multi-codebook diffusion large language model for omni-visual understanding and image generation. Unlike common unified models, it can also understand video and 3D content.
More from Multimodal
- GPT-6 generates a detailed Blender 3D model of Ilya Sutskever from a single photo — SIGKITTEN · 2026-09-05
- Astra 6 Creates 'The Universe Before Names': Art Spanning Anthrobots to Bell Inequalities — maluche · 2026-09-05
- Astra's 1000x1000 City Panoramas in Two Shots Prompt Analyst to Declare MC Bench Dead — Aizkmusic · 2026-09-05
- xAI Announces Imagine Odyssey Winners; $100K Top Prize Goes to All-Grok Film — XFreeze · 2026-09-05
- Dev tests fal.ai generative platform: streaming blocked by RTC issues, but Reactor is fun — flngr · 2026-09-05
- Clapper's AI stream exporter pairs with MiniMax H3 for real-time world rendering — flngr · 2026-09-05