Maxime Labonne demos d1: bounding boxes for named objects in a single forward pass
maximelabonne · x · 2026-10-28
Maxime Labonne showcased d1's grounding capability: you name the objects you care about and d1 detects and boxes every one of them in the same forward pass as answering other questions. He demoed it on his cat, noting bounding-box grounding makes it easy to fold object detection directly into multimodal Q&A without a separate detection model.
More from Models
- Frontier models hit 73.3% on SpatialBench 10 months after release, near 75% ceiling — kenbwork · 2026-10-28
- 5 models tested as Salesforce agents: Opus 87%, DeepSeek 82% at 1/75 the cost — TruthDeep8659 · 2026-10-28
- LiquidAI Ships 2B Omni Open Model With Day-One llama.cpp Support — ilintar · 2026-10-28
- Open-weight models now hold 72% of OpenRouter tokens as Tim Shi bets enterprises will own their intelligence — thursdai_pod · 2026-10-28
- EmbeddingGemma 2 runs real-time embeddings on a MacBook CPU alone: 20ms text, 100ms images — generativist · 2026-10-28
- Google Gemma rounds up EmbeddingGemma 2 community projects, from WebGPU photo search to Apple Silicon — GlennCameronjr · 2026-10-28