| M2-Encoder: Advancing Bilingual Image-Text Understanding by Large-scale Efficient Pretraining | ✓ Link | 10B | 88.5 | | M2-Encoder | 2024-01-29 |
| []() | | | 88.3 | | BASIC (Lion) | |
| CoCa: Contrastive Captioners are Image-Text Foundation Models | ✓ Link | | 86.3 | | CoCa | 2022-05-04 |
| Scaling Vision Transformers to 22 Billion Parameters | ✓ Link | | 85.9 | | LiT-22B | 2023-02-10 |
| Combined Scaling for Zero-shot Transfer Learning | | | 85.7 | | BASIC | 2021-11-19 |
| PaLI: A Jointly-Scaled Multilingual Language-Image Model | ✓ Link | | 85.4 | | LiT ViT-e | 2022-09-14 |
| LiT: Zero-Shot Transfer with Locked-image text Tuning | ✓ Link | | 84.5 | 75.7 | LiT-tuning | 2021-11-15 |
| Alternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perception | | | 83.9 | | IMP-MoE-L | 2023-05-10 |
| EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters | ✓ Link | | 83.8 | | EVA-CLIP-18B | 2024-02-06 |
| InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks | ✓ Link | | 83.2 | | InternVL-C | 2023-12-21 |
| The effectiveness of MAE pre-pretraining for billion-scale pretraining | ✓ Link | | 82.1 | | MAWS (ViT-2B) | 2023-03-23 |
| EVA-CLIP: Improved Training Techniques for CLIP at Scale | ✓ Link | | 82 | | EVA-CLIP-E/14+ | 2023-03-27 |
| []() | | | 81.8 | | CLIPA (ViT-H/14-336px) | |
| The effectiveness of MAE pre-pretraining for billion-scale pretraining | ✓ Link | | 81.1 | | MAWS (ViT-H) | 2023-03-23 |
| Learning Customized Visual Models with Retrieval-Augmented Knowledge | ✓ Link | | 78.5 | | REACT | 2023-01-17 |
| Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision | ✓ Link | | 76.4 | - | ALIGN | 2021-02-11 |
| Learning Transferable Visual Models From Natural Language Supervision | ✓ Link | | 76.2 | | CLIP（ViT-L/14-336px） | 2021-02-26 |
| AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities | ✓ Link | | 74.5 | | AltCLIP | 2022-11-12 |
| PaLI: A Jointly-Scaled Multilingual Language-Image Model | ✓ Link | | 72.11 | | PaLI | 2022-09-14 |
| Your Diffusion Model is Secretly a Zero-Shot Classifier | ✓ Link | | 61.4 | | Diffusion Classifier (zero-shot) | 2023-03-28 |
| Learning Transferable Visual Models From Natural Language Supervision | ✓ Link | | 59.6 | | CLIP (ResNet50) | 2021-02-26 |
| []() | | | | 76.5 | CWCL | |
| Learning Transferable Visual Models From Natural Language Supervision | ✓ Link | | | 31.3 | CLIP | 2021-02-26 |