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.

Training a FastFoundationStereo model v1.0.0

Use this resource to train a TAO FastFoundationStereo model for real-time stereo depth

New

  • 551 downloads
  • 203 likes
  • 166 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install real-time-stereo-depth-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 selects the FastFoundationStereo model, applies the distilled bp2 widths, writes the spec and runs training, evaluation and export. It lists the parameters to check and the seven-action deployment flow. You get a spec suited to real-time runs.

Mostly picked by

Updated November 2, 2026

Card for the “Training a FastFoundationStereo model” skill

Details

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

Similar resources

View all

What you get

  • references/spec-overrides.md: spec overrides
  • references/important-parameters.md: key parameters
  • references/error-patterns.md: common errors

Good fit for

  • You need real-time stereo depth
  • A full FoundationStereo is too slow for your target

Not for

    Tags

    • stereo
    • real time
    • disparity
    • TAO

    About

    You say

    “Train a FastFoundationStereo on my stereo pairs and export the distilled variant for real-time use.”

    Card for the “Training a FastFoundationStereo model” 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 TAO FastFoundationStereo model for real-time stereo depth
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md7.4 KB
    • config/skillspector-baseline.yaml1 KB
    • evals/evals.json0.8 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/error-patterns.md3.3 KB
    • references/important-parameters.md7.4 KB
    • references/parent-model-inference.md3 KB
    • references/setup-and-run.md5.6 KB
    • references/skill_info.yaml4.6 KB
    • references/spec_template_deploy.yaml2.7 KB
    • references/spec_template_evaluate.yaml1.9 KB
    • references/spec_template_export.yaml1.9 KB
    • references/spec_template_inference.yaml1.9 KB
    • references/spec_template_train.yaml2.7 KB
    • references/spec-overrides.md3.5 KB
    • references/tao-deploy-fast-foundation-stereo.md19.6 KB
    • references/tao-deploy-fast-foundation-stereo.skill_info.yaml3.7 KB
    • schemas/train.schema.json102.2 KB
    • skill-card.md4.8 KB
    • SKILL.md11.7 KB
    • skill.oms.sig8.2 KB

    Origin and licence

    Listed unmodified

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

    Notice

    « tao-train-fast-foundation-stereo » (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.