metadata
library_name: coreml
license: other
license_name: apple-ascl
license_link: https://github.com/apple/ml-mobileclip/blob/main/LICENSE_weights_data
datasets:
- apple/DataCompDR-1B
MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training
MobileCLIP was introduced in MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training (CVPR 2024), by Pavan Kumar Anasosalu Vasu, Hadi Pouransari, Fartash Faghri, Raviteja Vemulapalli, Oncel Tuzel.
This repository contains the text and image encoders of all variants of MobileCLIP exported to Core ML.
Highlights
- Our smallest variant
MobileCLIP-S0
obtains similar zero-shot performance as OpenAI's ViT-B/16 model while being 4.8x faster and 2.8x smaller. MobileCLIP-S2
obtains 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
Model | # Seen Samples (B) |
# Params (M) (img + txt) |
Latency (ms) (img + txt) |
IN-1k Zero-Shot Top-1 Acc. (%) |
Avg. Perf. (%) on 38 datasets |
---|---|---|---|---|---|
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 |
These Core ML models can be plugged-into the demo app provided in the official MobileCLIP repo