@qizhen2021/dsh-plugin-vision: A Vision Analysis Tool for DeepSeek Harness
DeepSeek Harness (DSH) enables text-only models to gain visual processing capabilities through the plugin `@qizhen2021/dsh-plugin-vision`. The plugin registers a `see` tool that provides three analysis channels: offline OCR (supporting macOS language detection), ASCII layout maps, and semantic descriptions (based on the `mimo-v2.5` model). Installation requires modifying `package.json` via Profile Bundles and configuring parameters such as the key path and timeout in `cordis.patch.yml`. When invoking the `see` tool, parameters can be specified to select the processing channel, and a structured JSON report is returned. Note that OCR is limited to macOS, large images require preprocessing, and the plugin's first
Read Moredsh-plugin-vision: DeepSeek Harness Vision Assistant Plugin
The `dsh-plugin-vision` plugin for DeepSeek Harness (DSH) addresses the lack of vision capabilities in the official API by registering the `vision_describe` tool. It enables the main model to call any OpenAI-compatible vision API to describe images. Installation must be performed using the official script, and the plugin supports configuration of custom API endpoints, API keys, and caching strategies. It relies on the global dependency tree, so the `dsh plugin add` command must not be used during installation. Additionally, when used with DSH Desktop, it can automatically convert pasted images to file paths, provided that dependencies are installed correctly.
Read Moredsh-plugin-vision: Enabling Gemini/GLM vision capabilities for DSH text models
dsh-plugin-vision is a visual capability plugin for DeepSeek Harness (DSH), designed to supplement local image understanding capabilities for pure text large models. This plugin does not require the main model to possess image reading capabilities; instead, it completes image description, OCR, and visual Q&A by calling the free visual APIs of Gemini or GLM and returns the results to the DSH session. Core functions include: supporting local image analysis for formats like PNG and JPG (single image limit of 20MB); providing the `see_image` tool for custom questions; supporting both Gemini and GLM providers with automatic routing, failover, and rate-limiting retry mechanisms; automatically compressing images larger than 4MB in Windows environments to reduce upload burden; and providing `vision_`
Read More