Qwen-MM-Plugins

wpnews.pro

English · 中文

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

🧩 Capabilities 🏗 Architecture 📦 Installation 🔧 Dependencies 🔑 Configuration 🚀 Quick Start 🧪 Development

Each capability is installed separately — a skill (so the model knows the toolset exists) plus an optional MCP server (the tools themselves).

We ship cookbooks of Qwen3.8-Max + these plugins in action — each capability's cookbook (linked in the table below) has its full tool listing, setup, and worked cases. Enjoy!

Capability	What it does	Install name	Cookbook
core
Foundational vision: dynamic-resolution reading of images / videos / documents / 3D models, plus OCR, grounding, segmentation, ASR, vision chat, and web search	`qwen-mm-plugins-core`

video-memoryqwen-mm-plugins-video-memory

TBDvideo-editqwen-mm-plugins-video-edit

TBDblender****running Blender via Python (thin client, 22 tools) — modeling / materials / lighting / renderingqwen-mm-plugins-blender

TBDfreecad****running FreeCAD (thin client, 14 tools) — modeling, property edits, STEP/STL import/export, FEM analysisqwen-mm-plugins-freecad

TBDedu-agent****skill-only, no MCP server)qwen-mm-plugins-edu-agent

TBDA capability = a skill (so the model knows the tools exist) + an optional MCP server (the tools themselves, launched on demand by uvx

— needs uv, no manual pip).

One script handles install · configure · verify · uninstall across every harness it supports (Claude Code · Codex · Qoder · OpenClaw · Qwen Code · Gemini CLI). It drives each harness's own native install under the hood — nothing reinvented — and writes a single shared config file (~/.qwen-mm-plugins/config

) that GUI and terminal harnesses both read, so you set things up once:

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

Or run one action at a time — bash install.sh install

/ configure

/ verify

/ uninstall

(what configure

and verify

do is detailed under Configuration and Dependencies).

Windows x64: use WSL2 (Ubuntu recommended) and clone the repository inside your WSL home directory (for example ~/code

), rather than under a mounted Windows drive such as /mnt/c

. Then run the same commands there. WSL2 is currently the only supported Windows environment; native Windows has not yet been validated. See the concise Windows notes.

Prefer your harness's own commands — or you're on opencode / pi / QwenPaw, which the installer doesn't cover? Register the skill + MCP yourself.

Plugin-marketplace harnesses (Claude Code · Qoder · Codex · OpenClaw · Qwen Code) — add the marketplace, then install a capability (replace <cap>

with core

/ video-memory

/ video-edit

/ blender

/ freecad

):

claude   plugin  marketplace add https://github.com/QwenLM/Qwen-MM-Plugins.git
claude   plugin  install       qwen-mm-plugins-<cap>@qwen-mm-plugins
qodercli plugins marketplace add https://github.com/QwenLM/Qwen-MM-Plugins.git
qodercli plugins install       qwen-mm-plugins-<cap>@qwen-mm-plugins
codex    plugin  marketplace add https://github.com/QwenLM/Qwen-MM-Plugins.git
codex    plugin  add           qwen-mm-plugins-<cap>@qwen-mm-plugins
openclaw plugins install       qwen-mm-plugins-<cap> --marketplace https://github.com/QwenLM/Qwen-MM-Plugins.git
qwen extensions install https://github.com/QwenLM/Qwen-MM-Plugins.git:qwen-mm-plugins-<cap> --consent

marketplace add

also accepts a local repo path; re-running is safe. On codex, marketplace add

does not refresh an already-added marketplace, so run codex plugin marketplace upgrade qwen-mm-plugins

before plugin add

to pick up newly-published capabilities.

Other harnesses (Gemini CLI · opencode · pi · QwenPaw · …) register the skill + MCP in their own config — exact per-harness blocks are in docs/en/installation.md. Easiest of all:

just ask the agent— "install

qwen-mm-plugins-<cap>

".uvx

installs the Python dependencies for the chosen profile on first launch — no manual pip. The only things you install yourself are system tools: ffmpeg

(video / audio), plus optional libreoffice

/ blender

/ texlive

/ chromium

for visualize

. Run bash install.sh verify

to self-test what's installed — it confirms your API key and reports any missing system tools (fetching each capability's env and running --check-system

under the hood). Full system-tool table, the edu-agent (skill-only) setup, and the blender/freecad thin-client notes: see docs/en/installation.md.

The API-based tools need a key — native image / video / document reading doesn't:

DASHSCOPE_API_KEY

—vision_chat

/ocr

/grounding

/transcribe_audio

/ generation / video-memory buildSERPER_API_KEY

—web_search

/web_extractor

/image_search

Export them in your shell, or persist them to ~/.qwen-mm-plugins/config

(read whenever a var isn't already in the environment — so GUI-launched harnesses pick them up too). The guided installer's Configure step writes that file for you:

bash install.sh configure

For non-interactive/automation setup and the full environment-variable catalog, see docs/en/installation.md.

Once a capability is installed, reference a file in your harness and just ask — the model picks the right tool automatically. Reading is dynamic-resolution: every image, video frame, and document page is auto-scaled to the VL model's patch grid, so a 4K screenshot's fine print and a tiny thumbnail both come in at the detail they need — no manual resizing.

@dashboard-4k.png      Read every number in this dashboard.
@report.pdf            Summarize page 3.
@receipt.jpg           OCR this and total the line items.
@street.jpg            Draw a box around every car in the scene.   # grounding

@lecture-2h.mp4        What are the main points, with timestamps?

                       Generate a 1024×1024 image of a red panda coding at night.
@/path/to/media        Help me edit this video down to about 3 minutes.

                       Model a low-poly wooden stool, add a warm key light, and render it.

                       Model an M6 hex bolt 30 mm long and export it as STEP.

@geometry-problem.png  Explain how to solve this as a narrated video.

See each capability's 🍳 cookbook for every tool, setup, and a worked case.

Development setup, contribution guidelines, and verification commands are in CONTRIBUTING.md. Detailed guides:

local development·

adding a capability·

testing.

Apache-2.0 — see LICENSE. The Blender and FreeCAD capabilities vendor third-party MIT-licensed code; see

and

src/capabilities/blender/NOTICE.md

for attribution.

src/capabilities/freecad/NOTICE.md

source & further reading

github.com — original article

Qwen-MM-Plugins

Run your AI side-project on zahid.host