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.

Onboarding a new architecture in AutoModel v1.0.0

Use this resource to onboard a new model architecture into NeMo AutoModel

New

  • 455 downloads
  • 109 likes
  • 137 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install onboard-new-model-architecture

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 classifies the architecture from config.json, names the files to create under components/models/, the registry update and the validation tests to add before using a full checkpoint. It follows the skill's five phases, from discovery to validation. You get a verifiable implementation plan.

Mostly picked by

Updated November 2, 2026

Card for the “Onboarding a new architecture in AutoModel” 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: onboarding phases
  • llm-patterns.md: code patterns for LLMs
  • vlm-patterns.md: code patterns for VLMs

Good fit for

  • You want to run a new Hugging Face model in AutoModel
  • You adapt an existing MoE or VLM to AutoModel

Not for

    Tags

    • AutoModel
    • architecture
    • onboarding
    • model

    About

    You say

    “Onboard this architecture described in config.json into AutoModel and list the tests to add.”

    Card for the “Onboarding a new architecture in AutoModel” 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 onboard a new model architecture into NeMo AutoModel
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md7.7 KB
    • capabilities-and-precision.md5 KB
    • evals/evals.json3.8 KB
    • LICENSE11.3 KB
    • llm-patterns.md12.2 KB
    • moe-patterns.md20 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • skill-card.md4.3 KB
    • SKILL.md22.8 KB
    • skill.oms.sig5.4 KB
    • vlm-patterns.md17 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/nemo-automodel-model-onboarding
    Licence
    Apache-2.0

    Notice

    « nemo-automodel-model-onboarding » (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.