sprainJinyu/dsh-vision-link

Route-preserving vision sidecar: a multimodal model looks at the image while your selected text model keeps reasoning and answering.

Sends attached images to a separately configured multimodal model, returns a text description, and leaves the chosen text model in control of reasoning and the final answer. It is designed as a lightweight sidecar that does not switch the session's active route.

Vision & Multimodal ★ 1
View on GitHub ↗

Install

dsh plugin --profile web add dsh-vision-link

The README documents the same command through the npx form; the plugin itself installs into the web profile.

Compatibility

Node.js >= 20.

Details

Alternatives

lijian-ui/dsh-vision-toggle · kaixinbaba/dsh-vision-recognizer · Einskyle/dsh-llm-vision-bridge

More plugins in Vision & Multimodal

Browse more in Vision & Multimodal

Guides for Vision & Multimodal plugins