convnext_base.dinov3_lvd1689m is a timm-style model identifier: a ConvNeXt-Base backbone with DINOv3 self-supervised weights pretrained on the LVD-1689M dataset. The model weights are hosted on Hugging Face, so that is the primary place to use it 2. Replicate hosts a related DINOv3 ViT-L distilled model trained on LVD-1689M 1, and Sentence Transformers covers embedding models generally 8.
- 1
Hugging Facehuggingface.coverified and signed
Hosts open models, datasets and libraries, with an SDK (huggingface-hub) and an MCP server to load and run vision models such as DINOv3 checkpoints 2.
official:MCP server · @huggingface/mcp-clientSDK · huggingface-hubCLI · text-embeddings-inferencerepository · pwc-cliskill · hf-cli+23 community tools built for it
for agents: huggingface-vision-trainer · pretrained-models · hf-mem
- 2
Replicatereplicate.comverified and signed
Runs open-source models through a cloud API; it hosts noah-bjorner/dinov3-vit-large-distilled, a DINOv3 ViT-L model trained on LVD-1689M 1.
+14 community tools built for it
for agents: noah-bjorner/dinov3-vit-large-distilled · lucataco/paligemma-3b-pt-224 · franz-biz/yolo-world-xl
- 3
Sentence Transformerssbert.netverified and signed
Python module for computing embeddings from text, images, audio and video with pretrained encoders, usable to load vision embedding backbones 8.
official:SDK · sentence-transformers+24 community tools built for it
for agents: XLNet · Networks · deep
Sources
- 1Replicate - Run AI with an APIreplicate.com · read 2026-10-29
- 2Hugging Face – The AI community building the future.huggingface.co · read 2026-10-29
- 8SentenceTransformers Documentation — Sentence Transformers documentationsbert.net · read 2026-10-29
For agents: the 10 actions behind this answer
noah-bjorner/dinov3-vit-large-distilled · noah-bjorner/dinov3-vit-large-distilled verified and signed
Meta's DINOv3 (ViT-L/16 Distilled). A state-of-the-art self-supervised vision transformer trained on the LVD-1689M dataset. Optimized for visual similarity, image retrieval, and dense feature extraction.
huggingface-vision-trainer · huggingface-vision-trainer verified and signed
Trains and fine-tunes vision models for object detection (D-FINE, RT-DETR v2, DETR, YOLOS), image classification (timm models — MobileNetV3, MobileViT, ResNet, ViT/DINOv3 — plus any Transformers classifier), and SAM/SAM2 segmentation using Hugging Face Transformers on Hugging Face Jobs cloud GPUs. C
wandb-primary · wandb-primary verified and signed
Primary W&B skill for broad or mixed Weights & Biases work: project overviews, W&B runs and artifacts, Weave traces and evaluations, Reports, Workspaces, Automations, Sweeps, Launch, and Agent Lens. Use when the task spans multiple W&B surfaces or the user asks generally what is happening in a W&B p
export · export verified and signed
Export a PyTorch model to .pte format for ExecuTorch. Use when converting models, lowering to edge, or generating .pte files.
· ONNX website read, tools not yet checked
YOLOv3 · ultralytics verified and signed
· Keras 3 API documentation website read, tools not yet checked
Keras documentation: Keras 3 API documentation
XLNet · sentence-transformers verified and signed
· vision-memory-mcp website read, tools not yet checked
Persistent visual cache for LLM-driven software development. Caches screenshots using perceptual hashing, vector search, and AX trees to prevent token overhead and visual hallucination loops.
list_models · LocalAI (Activepieces) verified and signed
List Models — List the models installed on your LocalAI server.
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.