metadata
license: mit
https://github.com/baaivision/EVA/tree/master/EVA-CLIP
Model Card
EVA-01-CLIP Series (MIM teacher: OpenAI CLIP-Large)
| model name | total #params | training precision | training data | training batch size | gpus for training | IN-1K zero-shot top-1 | MSCOCO T2I R@5 | weight |
|---|---|---|---|---|---|---|---|---|
EVA01_CLIP_g_14_psz14_s11B |
1.1B | fp16 |
LAION-400M | 41K | 256 A100(40GB) | 78.5 | 68.5 | 🤗 HF link (2.2GB) |
EVA01_CLIP_g_14_plus_psz14_s11B |
1.3B | fp16 |
Merged-2B | 114K | 112 A100(40GB) | 79.3 | 74.0 | 🤗 HF link (2.7GB) |
EVA-02-CLIP Series (MIM teacher: EVA01_CLIP_g_14_psz14_s11B)
| model name | total #params | training precision | training data | training batch size | gpus for training | IN-1K zero-shot top-1 | MSCOCO T2I R@5 | weight |
|---|---|---|---|---|---|---|---|---|
EVA02_CLIP_B_psz16_s8B |
149M | fp16 |
Merged-2B | 131K | 64 A100(40GB) | 74.7 | 66.9 | 🤗 HF link (300MB) |
EVA02_CLIP_L_psz14_s4B |
428M | fp16 |
Merged-2B | 131K | 128 A100(40GB) | 79.8 | 71.2 | 🤗 HF link (856MB) |
EVA02_CLIP_L_336_psz14_s6B |
428M | fp16 |
Merged-2B | 61K | 128 A100(40GB) | 80.4 | 71.7 | 🤗 HF link (856MB) |
EVA02_CLIP_E_psz14_s4B.pt |
4.7B | fp16 |
LAION-2B | 144K | 144 A100(80GB) | 81.9 | 74.7 | 🤗 HF link (9.4GB) |
To construct Merged-2B, we merged 1.6 billion samples from LAION-2B dataset with 0.4 billion samples from COYO-700M.
To our knowledge, EVA-CLIP series are the most performant open-sourced CLIP models at all scales, evaluated via zero-shot classification performance, especially on mainstream classification benchmarks such as ImageNet along with its variants. For more details about EVA-CLIP, please refer to our paper (coming very soon).