ZDTaichu5.0-9B
A 9B-parameter multimodal foundation model that pairs a Qwen3.5 language backbone with a vision encoder for general visual understanding, spatial reasoning and agentic tool use.
ZDTaichu5.0-9B is a multimodal foundation model for general visual understanding, spatial reasoning, agentic tool use and embodied-AI research. It combines a Qwen3.5-9B language backbone with a C-RADIOv4-H vision encoder, and accepts text, one or more images, and video at any resolution.
It is designed for image, document, chart and diagram understanding and OCR, visual mathematics and knowledge-grounded visual question answering, fine-grained 2D relations, multi-view association, 3D scene understanding and perspective taking, multi-step and multi-turn tool use, and spatial perception and planning for vision-language-action and embodied-AI adaptation. Within 10B-scale general-purpose vision-language models compared in its release blog, it reports leading results on spatial reasoning benchmarks such as SparBench, ViewSpatial, MMSI-Bench and MindCube-tiny, and on agent benchmarks including TAU2-Bench and IFEval.
Weights are published on Hugging Face and ModelScope in standard, FP8, NVFP4 and DSpark variants.