SenseTime open-sources SenseNova U1.5 Lite, 8B multimodal model
An 8-billion-parameter open-weights model combining visual understanding, generation and editing, it outputs natively at 4K and improves identity preservation and spatial structure during editing. It scores 60.18 on Qwen-Image-Bench (up from U1's 47.14) and is available via GitHub, Hugging Face and ModelScope.
How this story unfolded
1 day · 1 report · 4 community posts · 5 of 6 shown
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Reduce RAG costs on Amazon Bedrock with query-aware compression
- AWS and Panasonic Avionics use agentic AI for aircraft IFEC diagnostics
- User's Claude memory system backfired; Claude said user was the bottleneck
- Claude users discuss when to choose Sonnet over Opus
- Building token-efficient multi-agent systems