Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Adapt a DINOv3 backbone without labels v1.0.0

Use this resource to adapt a public DINOv3 backbone to your domain from unlabelled images

New

  • 415 downloads
  • 193 likes
  • 184 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install dinov3-self-supervised-adaptation

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent writes the training spec, runs teacher-student self-distillation, then converts and exports the teacher checkpoint for downstream tasks. It covers high-resolution and multi-GPU settings.

Mostly picked by

Updated November 2, 2026

Card for the “Adapt a DINOv3 backbone without labels” skill

Details

Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • SKILL.md : procedure and essential fields
  • references/dinov3-recipes.md : training recipes
  • references/dinov3-method.md : training method

Good fit for

  • You have many unlabelled images from your domain
  • You want an adapted backbone before training a detector

Not for

    Tags

    • dinov3
    • self-supervised
    • backbone
    • tao
    • adaptation

    About

    You say

    “Adapt a DINOv3 backbone to my industrial images, with no labels.”

    Card for the “Adapt a DINOv3 backbone without labels” skill

    Skill published by nvidia under the Apache-2.0 license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to adapt a public DINOv3 backbone to your domain from unlabelled images
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md7.2 KB
    • config/skillspector-baseline.yaml1 KB
    • evals/evals.json2.3 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/dinov3-method.md2.3 KB
    • references/dinov3-recipes.md3 KB
    • references/skill_info.yaml3.6 KB
    • references/spec_template_convert.yaml2.7 KB
    • references/spec_template_export.yaml2.6 KB
    • references/spec_template_inference.yaml2.4 KB
    • references/spec_template_train_highres.yaml3.4 KB
    • references/spec_template_train.yaml4 KB
    • schemas/convert.schema.json57.8 KB
    • schemas/export.schema.json58.7 KB
    • schemas/inference.schema.json58.8 KB
    • schemas/manifest.json20.5 KB
    • schemas/train.schema.json55.8 KB
    • skill-card.md4.8 KB
    • SKILL.md9 KB
    • skill.oms.sig7.8 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/tao-train-dinov3
    Licence
    Apache-2.0

    Notice

    « tao-train-dinov3 » (nvidia, Apache-2.0) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.