{"slug": "dinov2", "name": "DINOv2", "category": "ai-models", "category_name": "AI models", "description": "", "recommenders": 3, "commenters": 3, "warners": 0, "score": 2.435, "rising": -0.98, "first": "2025-11-19", "latest": "2026-01-06", "why": "We used DINOv2 as the backbone of our RF-DETR model", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 0, "compare": 0, "warn": 0}, "spark": [2, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "We used DINOv2 as the backbone of our RF-DETR model", "kind": "", "icon": true, "domain": "metademolab.com", "github_repo": null, "official_url": "https://dinov2.metademolab.com", "pricing": "", "aliases": ["DINOv2"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 744333185, "item_id": 46520824, "url": "https://news.ycombinator.com/item?id=46520824", "by": "Samin100", "date": "2026-01-06", "version": "2", "stance": "recommend", "quote": "Using a CLIP or Dino v2 model to produce image embeddings would probably improve the similarity search a lot", "excerpt": "This is awesome! Using a CLIP or Dino v2 model to produce image embeddings would probably improve the similarity search a lot - kind of similar to http://same.energy/ .", "context": "", "use_case": "image embeddings", "switched_from": "", "story": null}, {"id": 735822881, "item_id": 45988930, "url": "https://news.ycombinator.com/item?id=45988930", "by": "yeldarb", "date": "2025-11-19", "version": "2", "stance": "recommend", "quote": "We used DINOv2 as the backbone of our RF-DETR model", "excerpt": "We used DINOv2 as the backbone of our RF-DETR model, which is SOTA on realtime object detection and segmentation: https://github.com/roboflow/rf-detr It makes a great target to distill SAM3 to.", "context": "", "use_case": "model backbone", "switched_from": "", "story": "Meta Segment Anything Model 3"}, {"id": 735783297, "item_id": 45986456, "url": "https://news.ycombinator.com/item?id=45986456", "by": "sorenjan", "date": "2025-11-19", "version": "2", "stance": "recommend", "quote": "isn't their DINOv2 a better example of a large foundation model to distill from", "excerpt": "SAM3 is probably a great model to distill from when training smaller segmentation models, but isn't their DINOv2 a better example of a large foundation model to distill from for various computer vision tasks? I've seen it used for as starting point for models doing segmentation and depth estimation. Maybe there's a v3 coming soon? https://dinov2.metademolab.com/", "context": "", "use_case": "vision foundation model for distillation", "switched_from": "", "story": "Meta Segment Anything Model 3"}]}], "caveats": [], "who": [], "tags": [], "use_cases": ["image embeddings", "model backbone", "vision foundation model for distillation"], "compared": [], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=45982073", "title": "Meta Segment Anything Model 3", "mentions": 2}, {"url": "https://news.ycombinator.com/item?id=46518106", "title": "item 46518106", "mentions": 1}], "sentiment_by_month": {"2025-11": {"recommend": 2}, "2025-12": {}, "2026-01": {"recommend": 1}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "versions": [], "version_unnamed": 0, "quotes": [{"id": 744333185, "item_id": 46520824, "url": "https://news.ycombinator.com/item?id=46520824", "by": "Samin100", "date": "2026-01-06", "version": "2", "stance": "recommend", "quote": "Using a CLIP or Dino v2 model to produce image embeddings would probably improve the similarity search a lot", "excerpt": "This is awesome! Using a CLIP or Dino v2 model to produce image embeddings would probably improve the similarity search a lot - kind of similar to http://same.energy/ .", "context": "", "use_case": "image embeddings", "switched_from": "", "story": null}, {"id": 735822881, "item_id": 45988930, "url": "https://news.ycombinator.com/item?id=45988930", "by": "yeldarb", "date": "2025-11-19", "version": "2", "stance": "recommend", "quote": "We used DINOv2 as the backbone of our RF-DETR model", "excerpt": "We used DINOv2 as the backbone of our RF-DETR model, which is SOTA on realtime object detection and segmentation: https://github.com/roboflow/rf-detr It makes a great target to distill SAM3 to.", "context": "", "use_case": "model backbone", "switched_from": "", "story": "Meta Segment Anything Model 3"}, {"id": 735783297, "item_id": 45986456, "url": "https://news.ycombinator.com/item?id=45986456", "by": "sorenjan", "date": "2025-11-19", "version": "2", "stance": "recommend", "quote": "isn't their DINOv2 a better example of a large foundation model to distill from", "excerpt": "SAM3 is probably a great model to distill from when training smaller segmentation models, but isn't their DINOv2 a better example of a large foundation model to distill from for various computer vision tasks? I've seen it used for as starting point for models doing segmentation and depth estimation. Maybe there's a v3 coming soon? https://dinov2.metademolab.com/", "context": "", "use_case": "vision foundation model for distillation", "switched_from": "", "story": "Meta Segment Anything Model 3"}]}