15B (multimodal reasoning)
MIT
Open-weight multimodal reasoning model built on Phi-4-Reasoning with a SigLIP-2 vision encoder in a mid-fusion architecture; supports image+text input with text output, dynamic image resolution, and up to 3,600 visual tokens for detailed visual understanding; designed for both extended reasoning mode and direct non-reasoning mode, with model-card emphasis on OCR, GUI grounding, computer-use, and math/science reasoning tasks; reported context length is 16,384 tokens with strong benchmark results across AI2D, ChartQA, MathVista, MMMU, OCRBench, and ScreenSpot-v2.