From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models Paper • 2607.06553 • Published 8 days ago • 15
Generalization vs. Specialization: Evaluating Segment Anything Model (SAM3) Zero-Shot Segmentation Against Fine-Tuned YOLO Detectors Paper • 2512.11884 • Published Dec 9, 2025 • 2
Running Featured 92 LFM2.5-VL-1.6B WebGPU 🧠 92 In-browser vision-language inference with LFM2.5-VL-1.6B
Apple Neural Engine Collection Latest SOTA models supported on Apple Neural Engine • 7 items • Updated Dec 3, 2025 • 5
3DSRBench: A Comprehensive 3D Spatial Reasoning Benchmark Paper • 2412.07825 • Published Dec 10, 2024 • 12