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.

Knowledge distillation to compress an LLM v1.0.0

Use this resource to compress a large language model into a smaller one while keeping its performance

New

  • 545 downloads
  • 168 likes
  • 223 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install llm-knowledge-distillation-compression

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 sets up a distillation loop between a teacher model and a student model: loading both, the loss function and the training loop. You get the distillation code and a choice of strategy, logit loss or MiniLLM reverse KLD.

Mostly picked by

Updated November 2, 2026

Card for the “Knowledge distillation to compress an LLM” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: method, installation and training loop
  • references/minillm.md: MiniLLM distillation with reverse KLD

Good fit for

  • You want to cut the inference cost of a model without losing too much quality
  • You transfer a large model's answers to an open-source model

Not for

    Tags

    • distillation
    • compression
    • LLM
    • inference cost

    About

    You say

    “Distill my 70B teacher model into a 7B student model.”

    Card for the “Knowledge distillation to compress an LLM” skill

    Skill published by Orchestra Research under the MIT 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 compress a large language model into a smaller one while keeping its performance
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/minillm.md9.2 KB
    • SKILL.md13.7 KB

    Origin and licence

    Listed unmodified

    Author
    Orchestra Research
    Source
    github.com/orchestra-research/ai-research-skills/tree/773a52944ba4747a18bd4ae9ade53fff041adcbc/19-emerging-techniques/knowledge-distillation
    Licence
    MIT

    Notice

    « knowledge-distillation » (Orchestra Research, MIT) 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.