ImageNet Linear Probe
Unknown
Linear classification on frozen ImageNet-1K features. Used to evaluate representation quality of self-supervised and contrastive models without fine-tuning the backbone.
Benchmark Stats
SOTA History
top-1-accuracy
top-1-accuracy
Higher is better
| Rank | Model | Source | Score | Year | Paper |
|---|---|---|---|---|---|
| 1 | DINOv2 ViT-g/14 DINOv2 ViT-g/14, self-supervised via distillation. Linear probe on frozen features. Source: facebookresearch/dinov2 README pretrained models table. Paper: Oquab et al. 2023, arxiv:2304.07193. | Community | 86.5 | 2026 | Source |
| 2 | DINOv2 ViT-L/14 DINOv2 ViT-L/14, self-supervised via distillation. Linear probe on frozen features. Source: facebookresearch/dinov2 README pretrained models table. Paper: Oquab et al. 2023, arxiv:2304.07193. | Community | 86.3 | 2026 | Source |
| 3 | CLIP ViT-L/14 OpenAI CLIP ViT-L/14, contrastive pre-training on 400M image-text pairs. Linear probe on frozen features. 85.3% reported in original CLIP paper (Table 10, Appendix). Paper: Radford et al. 2021, arxiv:2103.00020. | Community | 85.3 | 2026 | Source |
| 4 | MAE ViT-H/14 Masked Autoencoder ViT-H/14. Linear probe on frozen features (PyTorch reimplementation). Source: facebookresearch/mae FINETUNE.md linear probing table. Paper: He et al. 2022, arxiv:2111.06377. Note: MAE is optimized for fine-tuning not linear probing; finetuned accuracy is 87.8%. | Community | 77.2 | 2026 | Source |
| 5 | MAE ViT-L/16 Masked Autoencoder ViT-L/16. Linear probe on frozen features (PyTorch reimplementation). Source: facebookresearch/mae FINETUNE.md. Paper: He et al. 2022, arxiv:2111.06377. | Community | 76 | 2026 | Source |