Trackio logbook, compact outputs, and artifact Bucket for reproducing Olaf-World (arXiv 2602.10104, OpenReview 5TiuerrwR8).
Niels Rogge
nielsr
AI & ML interests
Mainly interested in diving into complex Github repos and making AI easier and more accessible to everyone
Recent Activity
updated a bucket about 3 hours ago
nielsr/paperswithcode-backups new activity about 4 hours ago
Vickyinmyheart824/KeyFrame-Compass:Link paper, code and add citation new activity about 16 hours ago
thinkingmachines/Inkling:Add community evaluation results for AIME_2026, GPQA, HLE, MMMU_PRO, SWE-BENCH_PRO, SWE-BENCH_VERIFIEDOrganizations
![Unofficial Mistral Community [deprecated]'s profile picture](https://cdn-avatars.huggingface.co/v1/production/uploads/6141a88b3a0ec78603c9e784/HAdAzcs1CDB9OxgxT2W9K.png)
SigLIP improves upon CLIP with a sigmoid loss. Both English-only and multilingual checkpoints are released.
-
Sigmoid Loss for Language Image Pre-Training
Paper • 2303.15343 • Published • 12 -
google/siglip-base-patch16-224
Zero-Shot Image Classification • 0.2B • Updated • 1.55M • 87 -
google/siglip-base-patch16-256
Zero-Shot Image Classification • 0.2B • Updated • 35.7k • 6 -
google/siglip-base-patch16-384
Zero-Shot Image Classification • 0.2B • Updated • 25k • 11
DPT 3.1 (MiDaS) models, leveraging state-of-the-art vision backbones such as BEiT and Swinv2
-
Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer
Paper • 1907.01341 • Published • 1 -
Intel/dpt-beit-large-512
Depth Estimation • 0.3B • Updated • 724 • 9 -
Intel/dpt-beit-large-384
Depth Estimation • 0.3B • Updated • 375 -
Intel/dpt-beit-base-384
Depth Estimation • 0.1B • Updated • 332 • 1
Collection of image captioning models
-
Salesforce/blip-image-captioning-large
Image-to-Text • 0.5B • Updated • 794k • 1.48k -
microsoft/git-large-coco
Image-to-Text • 0.4B • Updated • 3.34k • 106 -
Salesforce/instructblip-vicuna-7b
Image-Text-to-Text • 8B • Updated • 10.6k • 102 -
Salesforce/blip2-flan-t5-xxl
Image-Text-to-Text • 12B • Updated • 1.15k • 94
DPT 3.0 (MiDaS) models, leveraging ViT and ViT-hybrid backbones
Depth Anything models, which are monocular depth estimation models trained on 62 million images
-
Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data
Paper • 2401.10891 • Published • 64 -
LiheYoung/depth-anything-large-hf
Depth Estimation • 0.3B • Updated • 419k • 65 -
LiheYoung/depth-anything-base-hf
Depth Estimation • 97.5M • Updated • 61.5k • 12 -
LiheYoung/depth-anything-small-hf
Depth Estimation • 24.8M • Updated • 25.1k • 35
Trackio logbook, compact outputs, and artifact Bucket for reproducing Olaf-World (arXiv 2602.10104, OpenReview 5TiuerrwR8).
Collection of image captioning models
-
Salesforce/blip-image-captioning-large
Image-to-Text • 0.5B • Updated • 794k • 1.48k -
microsoft/git-large-coco
Image-to-Text • 0.4B • Updated • 3.34k • 106 -
Salesforce/instructblip-vicuna-7b
Image-Text-to-Text • 8B • Updated • 10.6k • 102 -
Salesforce/blip2-flan-t5-xxl
Image-Text-to-Text • 12B • Updated • 1.15k • 94
SigLIP improves upon CLIP with a sigmoid loss. Both English-only and multilingual checkpoints are released.
-
Sigmoid Loss for Language Image Pre-Training
Paper • 2303.15343 • Published • 12 -
google/siglip-base-patch16-224
Zero-Shot Image Classification • 0.2B • Updated • 1.55M • 87 -
google/siglip-base-patch16-256
Zero-Shot Image Classification • 0.2B • Updated • 35.7k • 6 -
google/siglip-base-patch16-384
Zero-Shot Image Classification • 0.2B • Updated • 25k • 11
DPT 3.0 (MiDaS) models, leveraging ViT and ViT-hybrid backbones
DPT 3.1 (MiDaS) models, leveraging state-of-the-art vision backbones such as BEiT and Swinv2
-
Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer
Paper • 1907.01341 • Published • 1 -
Intel/dpt-beit-large-512
Depth Estimation • 0.3B • Updated • 724 • 9 -
Intel/dpt-beit-large-384
Depth Estimation • 0.3B • Updated • 375 -
Intel/dpt-beit-base-384
Depth Estimation • 0.1B • Updated • 332 • 1
Depth Anything models, which are monocular depth estimation models trained on 62 million images
-
Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data
Paper • 2401.10891 • Published • 64 -
LiheYoung/depth-anything-large-hf
Depth Estimation • 0.3B • Updated • 419k • 65 -
LiheYoung/depth-anything-base-hf
Depth Estimation • 97.5M • Updated • 61.5k • 12 -
LiheYoung/depth-anything-small-hf
Depth Estimation • 24.8M • Updated • 25.1k • 35