Models & Providers

Qwen-MM-Plugins

omdsh-dev/Qwen-MM-Plugins · ★ 0

omdsh-devCurated

Qwen multi-modal plugin support.

License
Apache-2.0
Added
2026-08-13
Maintenance
active
Install kind
github
Manifest status
Declared
Manifest path
package.json#dsh
Manifest checked
2026-08-19
Lifecycle scripts
None found

Static manifest checks report declarations only; they are not compatibility or security review.

安装

dsh plugin --profile web add github:omdsh-dev/Qwen-MM-Plugins

GitHub 源码安装:首次需按提示配置 allowBuilds 构建授权后重试

For install and environment setup, plus plugin development tutorials, see DSH community docs ↗

README

Source README · body: en · Read Chinese README

Table of contents
  1. Architecture
  2. Install
  3. DeepSeek Harness (DSH)
  4. Capabilities
  5. Try it
  6. Requirements and configuration
  7. Documentation
  8. License

English · 中文

Native multimodal plugins for Qwen models. Make any agent harness multimodal-native.

Architecture

Qwen-MM-Plugins architecture

Install

The guided installer supports Claude Code, CodeBuddy, Codex, Qoder, OpenClaw, Qwen Code, and Gemini CLI. Shared configuration lives in ~/.qwen-mm-plugins/config.

In-app setup for WorkBuddy, QoderWork, and QwenWork, plus manual setup for DeepSeek Harness, Hermes Agent, opencode, pi, and QwenPaw, is documented in the other harness guide.

curl -fsSL https://raw.githubusercontent.com/QwenLM/Qwen-MM-Plugins/main/install.sh | bash

Update the capabilities already installed in one harness:

curl -fsSL https://raw.githubusercontent.com/QwenLM/Qwen-MM-Plugins/main/install.sh | bash -s -- update

Released capabilities use independent, immutable tags. For local checkout installs, rollback, manual skill + MCP setup, dependencies, and Windows/WSL2, see the installation guide.

DeepSeek Harness (DSH)

The source repository includes a DSH carrier at packages/dsh. It mounts selected Qwen capabilities through the official @deepseek-ai/dsh-mcp-client; the Python MCP servers still come from this repository's uvx package source. Install the carrier in the target profile:

dsh plugin --profile web add @omdsh-dev/qwen-mm-plugins-dsh

The carrier requires the native-media DSH client line and relies on its built-in attachment admission and finalizeContent support; it does not carry a compatibility layer for older MCP clients. See the DSH setup guide for profile configuration and source-checkout mode.

Capabilities

Each capability is installed independently as a Skill plus an optional MCP server. Its install name is qwen-mm-plugins-<capability>.

Capability Use case Main requirements Cookbook
core Read images and video; visualize documents, code, data, 3D files, and more No API key; ffmpeg for audio/video; format-specific apps as needed Cookbook
api Qwen VL/Omni vision, OCR, grounding, ASR, segmentation, and audio-video understanding DashScope; ffmpeg for local audio/video Cookbook
search Web search, page extraction, and reverse-image search Serper, Exa, or Tavily key; image search requires Serper Cookbook
video-memory Build hierarchical memory for long-video QA DashScope; ffmpeg/ffprobe for builds Cookbook
video-edit Image, video, and audio generation with editing workflows DashScope; ffmpeg + Node/Chromium for full edits Cookbook
blender Model, texture, light, and render in Blender Blender; Xvfb on headless Linux Cookbook
freecad Parametric CAD, STEP/STL, and FEM workflows FreeCAD; CalculiX for FEM; Xvfb on headless Linux Cookbook
edu-agent Create Chinese math/science explainer videos and interactive pages Skill-only; Node/Chromium + ffmpeg; DashScope for narrated video Cookbook

Try it

After installing a capability, reference a file and ask naturally; the Skill selects the relevant MCP tool.

@report.pdf          Summarize page 3 and extract its table.
@meeting.mp4         Transcribe this with speaker labels and timestamps.
@place.jpg           Identify where this photo was taken and verify it on the web.
@lecture-2h.mp4      List the main points with timestamps.

core reads media at dynamic resolution, so manual resizing is normally unnecessary.

Requirements and configuration

  • uv provides uvx, which installs Python dependencies on demand.
  • Local core tools need no API key in the default native-image mode. Text-only caption fallback, cloud, and search capabilities need their provider credentials.
  • Video, document, browser, Blender, and FreeCAD workflows may need system applications.

Run the installer's Configure and Verify actions to set credentials and check dependencies. See Installation for prerequisites and the configuration reference for every setting.

Documentation

License

Apache-2.0 — see LICENSE. Third-party attribution for the Blender and FreeCAD integrations is recorded in their respective Blender and FreeCAD notices.