huggingface--transformers/docs/source/en/model_doc/metaclip_2.md

This model was contributed to Hugging Face Transformers on 2025-08-20.

MetaCLIP 2

Overview

MetaCLIP 2 is a replication of the original CLIP model trained on 300+ languages. It achieves state-of-the-art (SOTA) results on multilingual benchmarks (e.g., XM3600, CVQA, Babel‑ImageNet), surpassing previous SOTA such as mSigLIP and SigLIP‑2. The authors show that English and non-English worlds can mutually benefit and elevate each other.

This model was contributed by nielsr. The original code can be found here.

You can find all the MetaCLIP 2 checkpoints under the Meta organization.

Tip

Click on the MetaCLIP 2 models in the right sidebar for more examples of how to apply MetaCLIP 2 to different image and language tasks.

The example below demonstrates how to calculate similarity scores between multiple text descriptions and an image with [Pipeline] or the [AutoModel] class. Usage of the MetaCLIP 2 models is identical to the CLIP models, you just need the MetaClip2Model class instead of CLIPModel.

import torch

from transformers import pipeline


clip = pipeline(
   task="zero-shot-image-classification",
   model="facebook/metaclip-2-worldwide-huge-quickgelu",
   device=0
)
labels = ["a photo of a cat", "a photo of a dog", "a photo of a car"]
clip("http://images.cocodataset.org/val2017/000000039769.jpg", candidate_labels=labels)

import requests
from PIL import Image

from transformers import AutoModel, AutoProcessor


model = AutoModel.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu", attn_implementation="sdpa", device_map="auto")
processor = AutoProcessor.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")

url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
labels = ["a photo of a cat", "a photo of a dog", "a photo of a car"]

inputs = processor(text=labels, images=image, return_tensors="pt", padding=True).to(model.device)

outputs = model(**inputs)
logits_per_image = outputs.logits_per_image
probs = logits_per_image.softmax(dim=1)
most_likely_idx = probs.argmax(dim=1).item()
most_likely_label = labels[most_likely_idx]
print(f"Most likely label: {most_likely_label} with probability: {probs[0][most_likely_idx].item():.3f}")

MetaClip2Config

autodoc MetaClip2Config

MetaClip2TextConfig

autodoc MetaClip2TextConfig

MetaClip2VisionConfig

autodoc MetaClip2VisionConfig

MetaClip2Model

autodoc MetaClip2Model - forward - get_text_features - get_image_features

MetaClip2TextModel

autodoc MetaClip2TextModel - forward

MetaClip2TextModelWithProjection

autodoc MetaClip2TextModelWithProjection - forward

MetaClip2VisionModelWithProjection

autodoc MetaClip2VisionModelWithProjection - forward

MetaClip2VisionModel

autodoc MetaClip2VisionModel - forward

MetaClip2ForImageClassification

autodoc MetaClip2ForImageClassification - forward