timm/MobileCLIP2-S3-OpenCLIP
zero shot image classificationopen_clipopen_clipsafetensorsclipmobileclip2zero-shot-image-classificationarxiv:2508.20691apple-amlr
HuggingFace294.7K
Model card for MobileCLIP2-S3-OpenCLIP
These weights and model card are adapted from the original Apple model at https://huggingface.co/apple/MobileCLIP2-S3. This version uses canonical OpenCLIP configs and weight naming.
MobileCLIP2 was introduced in MobileCLIP2: Improving Multi-Modal Reinforced Training (TMLR August 2025 Featured), by Fartash Faghri, Pavan Kumar Anasosalu Vasu, Cem Koc, Vaishaal Shankar, Alexander T Toshev, Oncel Tuzel, Hadi Pouransari.
This repository contains the MobileCLIP2-S3 checkpoint.
Highlights
MobileCLIP2-S4matches the accuracy of SigLIP-SO400M/14 with 2x fewer parameters and surpasses DFN ViT-L/14 at 2.5x lower latency measured on iPhone12 Pro Max.MobileCLIP-S3/S4are our new architectures trained on MobileCLIP’s training dataset, DataCompDR-1B (dashed lines).- Our smallest variant
MobileCLIP-S0obtains similar zero-shot performance as OpenAI's ViT-B/16 model while being 4.8x faster and 2.8x smaller. MobileCLIP-S2obtains better avg zero-shot performance than SigLIP's ViT-B/16 model while being 2.3x faster and 2.1x smaller, and trained with 3x less seen samples.MobileCLIP-B (LT)attains zero-shot ImageNet performance of 77.2% which is significantly better than recent works like DFN and SigLIP with similar architectures or even OpenAI's ViT-L/14@336.
Checkpoints and Results (Original Apple links)
| Model | # Seen Samples (B) | # Params (M) (img + txt) | Latency (ms) (img + txt) | IN-1k Zero-Shot Top-1 Acc. (%) | Avg. Perf. (%) on 38 datasets |
|---|---|---|---|---|---|
| MobileCLIP2-S0 | 13 | 11.4 + 42.4 | 1.5 + 1.6 | 71.5 | 59.7 |
| MobileCLIP2-S2 | 13 | 35.7 + 63.4 | 3.6 + 3.3 | 77.2 | 64.1 |
| MobileCLIP2-B | 13 | 86.3 + 63.4 | 10.4 + 3.3 | 79.4 | 65.8 |
| MobileCLIP2-S3 | 13 | 125.1 + 123.6 | 8.0 + 6.6 | 80.7 | 66.8 |
| MobileCLIP2-L/14 | 13 | 304.3 + 123.6 | 57.9 + 6.6 | 81.9 | 67.8 |
| MobileCLIP2-S4 | 13 | 321.6 + 123.6 | 19.6 + 6.6 | 81.9 | 67.5 |
| MobileCLIP-S0 | 13 | 11.4 + 42.4 | 1.5 + 1.6 | 67.8 | 58.1 |
| MobileCLIP-S1 | 13 | 21.5 + 63.4 | 2.5 + 3.3 | 72.6 | 61.3 |
| MobileCLIP-S2 | 13 | 35.7 + 63.4 | 3.6 + 3.3 | 74.4 | 63.7 |
| MobileCLIP-B | 13 | 86.3 + 63.4 | 10.4 + 3.3 | 76.8 | 65.2 |
| MobileCLIP-B (LT) | 36 | 86.3 + 63.4 | 10.4 + 3.3 | 77.2 | 65.8 |
| MobileCLIP-S3 | 13 | 125.1 + 123.6 | 8.0 + 6.6 | 78.3 | 66.3 |
| MobileCLIP-L/14 | 13 | 304.3 + 123.6 | 57.9 + 6.6 | 79.5 | 66.9 |
| MobileCLIP-S4 | 13 | 321.6 + 123.6 | 19.6 + 6.6 | 79.4 | 68.1 |
How to Use
import torch
import open_clip
from PIL import Image
from urllib.request import urlopen
from timm.utils import reparameterize_model
model, _, preprocess = open_clip.create_model_and_transforms('MobileCLIP2-S3', pretrained='dfndr2b')
model.eval()
tokenizer = open_clip.get_tokenizer('MobileCLIP2-S3')
# For inference/model exporting purposes, optionally reparameterize for better performance
model = reparameterize_model(model)
image = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
image = preprocess(image).unsqueeze(0)
text = tokenizer(["a diagram", "a dog", "a cat", "a doughnut"])
with torch.no_grad(), torch.amp.autocast(image.device.type):
image_features = model.encode_image(image)
text_features = model.encode_text(text)
image_features /= image_features.norm(dim=-1, keepdim=True)
text_features /= text_features.norm(dim=-1, keepdim=True)
text_probs = (100.0 * image_features @ text_features.T).softmax(dim=-1)
print("Label probs:", text_probs)