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

PalateKit

Sign in
Menu

Request a resource

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

0 / 280

A page, a video or an image online.

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.

Train a real-time RT-DETR detector v1.0.0

Use this resource to train a low-latency 2D object detector and prepare it for deployment

New

  • 497 downloads
  • 162 likes
  • 178 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install real-time-detr-training

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 spec, runs training, then depending on your need, distillation, quantization or TensorRT export. It applies the default export settings and the supplied spec templates.

Mostly picked by

Updated November 2, 2026

Card for the “Train a real-time RT-DETR detector” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • SKILL.md : actions, requirements and default settings
  • references/spec_template_distill.yaml : distillation template
  • references/spec_template_quantize.yaml : quantization template

Good fit for

  • You want a fast detector for live video
  • You need to shrink an RT-DETR model by distillation before deployment

Not for

    Tags

    • rt-detr
    • real-time
    • distillation
    • tao
    • detector

    About

    You say

    “Train an RT-DETR, then distil it for faster inference.”

    Card for the “Train a real-time RT-DETR detector” 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 train a low-latency 2D object detector and prepare it for deployment
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md7.1 KB
    • config/skillspector-baseline.yaml0.8 KB
    • evals/evals.json0.8 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/skill_info.yaml7.3 KB
    • references/spec_template_deploy_evaluate.yaml0.5 KB
    • references/spec_template_deploy_gen_trt_engine.yaml0.7 KB
    • references/spec_template_deploy_inference.yaml0.6 KB
    • references/spec_template_distill.yaml4.3 KB
    • references/spec_template_evaluate.yaml3.2 KB
    • references/spec_template_export.yaml3.3 KB
    • references/spec_template_inference.yaml3.3 KB
    • references/spec_template_quantize.yaml3.1 KB
    • references/spec_template_train.yaml3.1 KB
    • references/tao-deploy-rtdetr.md6.2 KB
    • references/tao-deploy-rtdetr.skill_info.yaml3.4 KB
    • schemas/distill.schema.json52.2 KB
    • schemas/evaluate.schema.json55.9 KB
    • schemas/export.schema.json56.5 KB
    • schemas/inference.schema.json56.8 KB
    • schemas/manifest.json18.6 KB
    • schemas/quantize.schema.json52.2 KB
    • schemas/train.schema.json52.2 KB
    • skill-card.md4.2 KB
    • SKILL.md17.7 KB
    • skill.oms.sig9.4 KB

    Origin and licence

    Listed unmodified

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

    Notice

    « tao-train-rtdetr » (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.