Instructions to use second-state/Qwen2-VL-2B-Instruct-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use second-state/Qwen2-VL-2B-Instruct-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="second-state/Qwen2-VL-2B-Instruct-GGUF")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("second-state/Qwen2-VL-2B-Instruct-GGUF")
model = AutoModelForMultimodalLM.from_pretrained("second-state/Qwen2-VL-2B-Instruct-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use second-state/Qwen2-VL-2B-Instruct-GGUF 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 second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf second-state/Qwen2-VL-2B-Instruct-GGUF: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 second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf second-state/Qwen2-VL-2B-Instruct-GGUF: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 second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M

Use Docker

docker model run hf.co/second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use second-state/Qwen2-VL-2B-Instruct-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "second-state/Qwen2-VL-2B-Instruct-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "second-state/Qwen2-VL-2B-Instruct-GGUF",
		"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/second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M

SGLang

How to use second-state/Qwen2-VL-2B-Instruct-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "second-state/Qwen2-VL-2B-Instruct-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "second-state/Qwen2-VL-2B-Instruct-GGUF",
		"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 images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "second-state/Qwen2-VL-2B-Instruct-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "second-state/Qwen2-VL-2B-Instruct-GGUF",
		"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"
						}
					}
				]
			}
		]
	}'

Ollama
How to use second-state/Qwen2-VL-2B-Instruct-GGUF with Ollama:
```
ollama run hf.co/second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M
```

Unsloth Studio

How to use second-state/Qwen2-VL-2B-Instruct-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for second-state/Qwen2-VL-2B-Instruct-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for second-state/Qwen2-VL-2B-Instruct-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for second-state/Qwen2-VL-2B-Instruct-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use second-state/Qwen2-VL-2B-Instruct-GGUF with Docker Model Runner:
```
docker model run hf.co/second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M
```

Lemonade

How to use second-state/Qwen2-VL-2B-Instruct-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull second-state/Qwen2-VL-2B-Instruct-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen2-VL-2B-Instruct-GGUF-Q4_K_M

List all available models

lemonade list

Qwen2-VL-2B-Instruct-GGUF

Original Model

Qwen/Qwen2-VL-2B-Instruct

Run with LlamaEdge

LlamaEdge version: v0.16.0

Prompt template

Prompt type: qwen2-vision

Prompt string

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
<|vision_start|>{image_placeholder}<|vision_end|>{user_prompt}<|im_end|>
<|im_start|>assistant

Context size: 32000

Run as LlamaEdge service

wasmedge --dir .:. \
  --nn-preload default:GGML:AUTO:Qwen2-VL-2B-Instruct-Q5_K_M.gguf \
  llama-api-server.wasm \
  --model-name Qwen2-VL-2B-Instruct \
  --prompt-template qwen2-vision \
  --llava-mmproj Qwen2-VL-2B-Instruct-vision-encoder.gguf \
  --ctx-size 32000

Quantized GGUF Models

Name	Quant method	Bits	Size	Use case
Qwen2-VL-2B-Instruct-Q2_K.gguf	Q2_K	2	676 MB	smallest, significant quality loss - not recommended for most purposes
Qwen2-VL-2B-Instruct-Q3_K_L.gguf	Q3_K_L	3	880 MB	small, substantial quality loss
Qwen2-VL-2B-Instruct-Q3_K_M.gguf	Q3_K_M	3	824 MB	very small, high quality loss
Qwen2-VL-2B-Instruct-Q3_K_S.gguf	Q3_K_S	3	761 MB	very small, high quality loss
Qwen2-VL-2B-Instruct-Q4_0.gguf	Q4_0	4	935 MB	legacy; small, very high quality loss - prefer using Q3_K_M
Qwen2-VL-2B-Instruct-Q4_K_M.gguf	Q4_K_M	4	986 MB	medium, balanced quality - recommended
Qwen2-VL-2B-Instruct-Q4_K_S.gguf	Q4_K_S	4	940 MB	small, greater quality loss
Qwen2-VL-2B-Instruct-Q5_0.gguf	Q5_0	5	1.10 GB	legacy; medium, balanced quality - prefer using Q4_K_M
Qwen2-VL-2B-Instruct-Q5_K_M.gguf	Q5_K_M	5	1.13 GB	large, very low quality loss - recommended
Qwen2-VL-2B-Instruct-Q5_K_S.gguf	Q5_K_S	5	1.10 GB	large, low quality loss - recommended
Qwen2-VL-2B-Instruct-Q6_K.gguf	Q6_K	6	1.27 GB	very large, extremely low quality loss
Qwen2-VL-2B-Instruct-Q8_0.gguf	Q8_0	8	1.65 GB	very large, extremely low quality loss - not recommended
Qwen2-VL-2B-Instruct-f16.gguf	f16	16	3.09 GB
Qwen2-VL-2B-Instruct-vision-encoder.gguf	f16	16	2.66 GB