multimodal-ai

Image understanding, OCR, and persistent visual evidence for text-only DeepSeek Harness models

给 DeepSeek 安装一双眼睛和一支画笔:会话里直接贴截图/图片,GLM 视觉模型先精确转写图片内容(报错信息、代码、界面逐字保留),然后 DeepSeek 继续处理你的问题——同一轮完成,全程无感;需要配图时,DeepSeek 自动调用文生图后端出图并显示在会话中。

dsh全模态工作站插件，让模型支持视频、图片、语音的输入与输出，支持comfyui图像生成工具调用。Any-to-Any.

AdaTile-MCP: high-resolution image adaptive tiling MCP server for DeepSeek vision model (deepseek-v4-flash-vision-exp). L1-L6 pipeline (fastpath, saliency, adaptive tiling, Files API assembly, streaming VLM, rule-based merge) + eval harness. Setup: clone, run setup.bat, add your agent MCP config -> see README.

AI-powered PDF reader for DeepSeek Harness with annotations, multi-PDF workflows, mixed image-text evidence, and on-demand OCR.