Instructions to use microsoft/Mage-VL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use microsoft/Mage-VL with Transformers:

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

pipe = pipeline("image-text-to-text", model="microsoft/Mage-VL", trust_remote_code=True)
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 AutoModelForImageTextToText
model = AutoModelForImageTextToText.from_pretrained("microsoft/Mage-VL", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use microsoft/Mage-VL with vLLM:

Install from pip and serve model

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

SGLang

How to use microsoft/Mage-VL 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 "microsoft/Mage-VL" \
    --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": "microsoft/Mage-VL",
		"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 "microsoft/Mage-VL" \
        --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": "microsoft/Mage-VL",
		"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"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use microsoft/Mage-VL with Docker Model Runner:
```
docker model run hf.co/microsoft/Mage-VL
```

Mage-VL

File size: 2,456 Bytes

12acbba

// Copyright (c) Microsoft Corporation.
// Licensed under the MIT License.

#pragma once
#include "rans.h"
#include <memory>

#include <pybind11/numpy.h>
#include <pybind11/pybind11.h>
#include <pybind11/stl.h>

namespace py = pybind11;

// the classes in this file only perform the type conversion
// from python type (numpy) to C++ type (vector)
class RansEncoder {
public:
    RansEncoder();

    RansEncoder(const RansEncoder&) = delete;
    RansEncoder(RansEncoder&&) = delete;
    RansEncoder& operator=(const RansEncoder&) = delete;
    RansEncoder& operator=(RansEncoder&&) = delete;

    void encode_y(const py::array_t<int16_t>& symbols, const int cdf_group_index);
    void encode_z(const py::array_t<int8_t>& symbols, const int cdf_group_index,
                  const int start_offset, const int per_channel_size);
    void flush();
    py::array_t<uint8_t> get_encoded_stream();
    void reset();
    int add_cdf(const py::array_t<int32_t>& cdfs, const py::array_t<int32_t>& cdfs_sizes,
                const py::array_t<int32_t>& offsets);
    void empty_cdf_buffer();
    void set_use_two_encoders(bool b);
    bool get_use_two_encoders();

private:
    std::shared_ptr<RansEncoderLib> m_encoder0;
    std::shared_ptr<RansEncoderLib> m_encoder1;
    bool m_use_two_encoders{ false };
};

class RansDecoder {
public:
    RansDecoder();

    RansDecoder(const RansDecoder&) = delete;
    RansDecoder(RansDecoder&&) = delete;
    RansDecoder& operator=(const RansDecoder&) = delete;
    RansDecoder& operator=(RansDecoder&&) = delete;

    void set_stream(const py::array_t<uint8_t>&);

    void decode_y(const py::array_t<uint8_t>& indexes, const int cdf_group_index);
    py::array_t<int8_t> decode_and_get_y(const py::array_t<uint8_t>& indexes, const int cdf_group_index);
    void decode_z(const int total_size, const int cdf_group_index, const int start_offset,
                  const int per_channel_size);
    py::array_t<int8_t> get_decoded_tensor();
    int add_cdf(const py::array_t<int32_t>& cdfs, const py::array_t<int32_t>& cdfs_sizes,
                const py::array_t<int32_t>& offsets);
    void empty_cdf_buffer();
    void set_use_two_decoders(bool b);
    bool get_use_two_decoders();

private:
    std::shared_ptr<RansDecoderLib> m_decoder0;
    std::shared_ptr<RansDecoderLib> m_decoder1;
    bool m_use_two_decoders{ false };
};

std::vector<uint32_t> pmf_to_quantized_cdf(const std::vector<float>& pmf, int precision);