SenseLogic releases SenseNova U1.5 Lite: 8B open-source model with native 4K output and precise editing
量子位 · wechat · 2026-08-21
SenseLogic has open-sourced the 8B parameter image model SenseNova U1.5 Lite, featuring a unified native architecture. It supports complex instruction following (3-4k characters) and native 4K high-res output, showing stability in tasks like multi-image collage, complex typography, and precise local editing (e.g., modifying poster text while preserving texture). Technically, it uses a 'split then merge' strategy, training individual Experts and distilling them back via MOPD, enabling visual understanding and generation to reinforce each other for a complete workflow from generation to delivery.
More from Multimodal
- Opus 5.5 + Midjourney: 'models have taste, you just point them right' — yuwen_lu_ · 2026-10-16
- Envato Launches AI Creative Toolkit Bundling Nano Banana, Veo and More — alifcoder · 2026-10-16
- Musk amplifies Grok Imagine dance video demo tackling generative AI's hardest motion — elonmusk · 2026-10-16
- CameraVLM: 9B model cracks camera-motion understanding where most VLMs flip coins — tencent · 2026-10-16
- Lluna: graph-based unimodal collaboration fixes missing-modality inference failures — williammary · 2026-10-16
- 'Game of Malibu': stunning AI-generated beach concert scene hits the 3-minute generation limit — bmallCakeDiver · 2026-10-16