NUS Jeveration generates 512px images in one forward pass, 78ms on H200, 122x faster
multimodalart · x · 2026-11-19
Jeveration from NUS applies the 'System One' approach to image generation: a frozen vision-language model outputs an entire 512px image in a single forward pass.
- 78ms per image on one H200, up to 122x faster than unified generation models
- One checkpoint covers three tasks
- Quality is far from SOTA, but generation is astonishingly fast
Demo images (a sunflower still life, a cat holding a camera) look plausible. multimodalart: quality is far from SOTA, but the speed is insane.
Related event: Jeveration: Frozen VLM Generates 512px Images in One Forward Pass, 78ms(5 posts)→
More from Multimodal
- Building a Claude-controlled video editor over MCP: lessons from ~200 tools, approvals and context bloat — MrxAnomynous · 2026-11-20
- OpenMontage: a self-hosted pipeline for consistent AI short dramas — Direct_Grab7063 · 2026-11-20
- CapCut launches ChatGPT video plugin: generate and edit timeline-based videos in chat — alifcoder · 2026-11-20
- I gave Claude a video editor — it made a 21-shot short I can fix shot-by-shot — MrxAnomynous · 2026-11-20
- Demi Guo previews real-time AI video model for directing films like a set — ying11231 · 2026-11-20
- Full automated GTA 6 content pipeline: Apify + ElevenLabs + Claude + Codex — eptwts · 2026-11-20