Deepseek Vision Bridge

Deploy and maintain image understanding (OCR + local VLM + cloud VL) and image generation for Codex connected to text-only models like DeepSeek. Use when the user wants to see/read images in the conversation bar with a text-only model, set up automatic OCR or vision-model processing for pasted images, configure image generation via SiliconFlow or local Stable Diffusion/Flux, fix a broken vision proxy, or asks "看图/生图/贴图识别/图片理解/OCR代理" on a Codex+DeepSeek setup. Deploys a local proxy between Codex and the model that converts images to text before forwarding.

aiskillstore Updated

File contents

aiskillstore/marketplace/tree/main/skills/lightlw/deepseek-vision-bridge commit eaed2b5b5e

Frequently asked questions

npx skillmds@latest add aiskillstore/deepseek-vision-bridge