Qwen3-VL-Embedding-2B-GGUF-vecBox

Instructions to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Use Docker

docker model run hf.co/alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

LM Studio
Jan

vLLM

How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Ollama
How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with Ollama:
```
ollama run hf.co/alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M
```
Unsloth Desktop

How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with Docker Model Runner:
```
docker model run hf.co/alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M
```

Lemonade

How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-VL-Embedding-2B-GGUF-vecBox-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "alpaim/Qwen3-VL-Embedding-2B-GGUF-vecBox:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Qwen3-VL-Embedding-2B-GGUF for vecBox

Not intended for general use! This repository is strictly maintained as a dependency for the vecBox. The file structure, metadata, and configurations may be modified in the future without notice to fit internal requirements.

If you are looking to run this model using standard software (like llama.cpp), please use the original repositories instead:

For GGUF quantized weights: DevQuasar/Qwen.Qwen3-VL-Embedding-2B-GGUF

For the original model: Qwen/Qwen3-VL-Embedding-2B

This is a custom repack of the quantized model DevQuasar/Qwen.Qwen3-VL-Embedding-2B-GGUF specifically packaged for the vecBox.

What's different?

This repository contains the exact same GGUF weights from the original DevQuasar repository. However, it additionally includes essential metadata files (config.json, tokenizer.json, etc.) extracted from the original unquantized model. These files are required for direct loading and tokenization mapping via the Rust candle framework inside the vecBox.