---
title: medsiglip-448
canonical_url: "https://www.modelscope.cn/models/mlx-community/medsiglip-448"
md_url: "https://www.modelscope.cn/models/mlx-community/medsiglip-448.md"
repository: mlx-community/medsiglip-448
last_updated: 2026-01-22
license: other
model_type:
  - siglip
architectures:
  - SiglipModel
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 24
stars: 0
tags:
  - vision
  - medical
  - radiology
  - dermatology
  - pathology
  - ophthalmology
  - chest-x-ray
  - mlx
---

# medsiglip-448

> medsiglip-448 - mlx-community 在 ModelScope 开源的模型。mlx-community/medsiglip-448

mlx-community/medsiglip-448 是 ModelScope 魔搭社区上的机器学习模型，采用 other 许可。

- **Repository**: mlx-community/medsiglip-448
- **License**: other
- **Tags**: vision, medical, radiology, dermatology, pathology, ophthalmology, chest-x-ray, mlx
- **Downloads**: 24
- **Stars**: 0
- **Last updated**: 2026-01-22

Source: https://www.modelscope.cn/models/mlx-community/medsiglip-448

---

# mlx-community/medsiglip-448

The Model [mlx-community/medsiglip-448](https://huggingface.co/mlx-community/medsiglip-448) was converted to MLX format from [google/medsiglip-448](https://huggingface.co/google/medsiglip-448) using mlx-lm version **0.0.6**.

## Use with mlx

```bash
pip install mlx-embeddings
```

```python
from mlx_embeddings import load, generate
import mlx.core as mx

model, tokenizer = load("mlx-community/medsiglip-448")

# For image-text embeddings
images = [
    "./images/cats.jpg",  # cats
]
texts = ["a photo of cats", "a photo of a desktop setup", "a photo of a person"]

# Process all image-text pairs
outputs = generate(model, processor, texts, images=images)
logits_per_image = outputs.logits_per_image
probs = mx.sigmoid(logits_per_image) # probabilities for this image
for i, image in enumerate(images):
    print(f"Image {i+1}:")
    for j, text in enumerate(texts):
        print(f"  {probs[i][j]:.1%} match with '{text}'")
    print()


```