Zero-shot Image Classification on ImageNet 1k and Six Robustness Variants (test)
65.58Zero-shot Mean AccuracyTuneCLIP
Evaluation results
| Method | Links | |
|---|---|---|
| TuneCLIPBase Model=SigLIP ViT-B/16, WCT. (hrs)=9.272026.01 | 65.58 | |
| BaselineBase Model=SigLIP ViT-B/16, WCT. (hrs)=N/A2026.01 | 63.12 | |
| TuneCLIPBase Model=OpenAI ViT-B/16, WCT. (hrs)=8.622026.01 | 59.36 | |
| BaselineBase Model=OpenAI ViT-B/16, WCT. (hrs)=N/A2026.01 | 57.67 | |
| OpenCLIPBase Model=OpenAI ViT-B/16, WCT. (hrs)=5.462026.01 | 54.99 | |
| FastCLIPBase Model=OpenAI ViT-B/16, WCT. (hrs)=4.212026.01 | 54.57 | |
| OpenCLIPBase Model=SigLIP ViT-B/16, WCT. (hrs)=7.552026.01 | 40.21 | |
| FastCLIPBase Model=SigLIP ViT-B/16, WCT. (hrs)=4.282026.01 | 39.22 |