GrassVison

What it does

Add image understanding to text-only LLMs like DeepSeek seamlessly. Provides OpenAI-compatible API, automatically sends image requests to vision model, injects structured results into text model, making enhanced model experience close to native multimodal.

Add image understanding to text-only LLMs like DeepSeek seamlessly. Provides OpenAI-compatible API, automatically sends image requests to vision model, injects structured results into text model, making enhanced model experience close to native multimodal. Not yet verified — install and test it yourself.

“Untested” means our automated CI has not yet installed this plugin. Feature descriptions and version compatibility are the author’s claims. This is not a security audit and not an endorsement of third-party code.

Install

This plugin is GitHub source (not published to npm) — install it straight from the repo:

dsh plugin --profile web add github:moduqishi/GrassVison

1. Install pnpm once: npm install -g pnpm.
2. Add it from GitHub: dsh plugin --profile web add github:moduqishi/GrassVison.
3. Restart dsh web.

Note: plugins that ship a prepare build script may prompt pnpm to allow the build — add the key it prints under allowBuilds in pnpm-workspace.yaml, then re-run.

Test report

Not yet verified — install and test it yourself.

Repository

github.com/moduqishi/GrassVison ↗

Share this badge

Paste this into your README — it links back to this page and updates automatically when our test status changes:

[![dshbase listed](https://dshbase.com/badges/GrassVison.svg)](https://dshbase.com/plugins/GrassVison/)

Copy badge markdown

More in Vision

Browse all 1912 plugins →