Integrations / Unsloth

RunDocumented

Fine-tune LLMs with Unsloth

Run Unsloth LoRA or QLoRA fine-tuning as a finite job and download the adapter.

The steps follow the tool's documented settings. No tested run is recorded for this page yet.

Why use Badgr for Unsloth

Badgr interface
Run
Where Unsloth fits
Training runtime
Why Badgr
Efficient LLM fine-tuning
  • A built-in template starts LoRA or QLoRA fine-tuning without environment setup.
  • A finite job with a cost cap, billing only while it trains.
  • The adapter comes back as output files, with a receipt.

What you can do

  • LoRA or QLoRA fine-tune of an open model on your data
  • Train a small adapter and merge it later
  • Try several datasets as separate capped jobs

Quickstart: Start the Unsloth template

badgr template info unsloth
badgr run template unsloth

How it works

  1. 1.You run one command
  2. 2.Badgr checks it and picks a GPU
  3. 3.Your code runs and logs stream back
  4. 4.Outputs are saved and billing stops

What you get

  • Logs while it runs
  • Output files, downloaded with badgr artifacts <id>
  • A receipt with cost and duration
  • LoRA adapter files

Requirements

  • The Badgr CLI installed and logged in (npm install -g badgr-cli)
  • A funded Badgr balance
  • A dataset in a format your training script reads
  • Hub access for gated base models

What it costs

  • A job is billed for the time it runs and stops billing when your command exits.
  • Pass --max-cost to set the most a job can spend. Badgr stops it if it reaches the cap.
  • Badgr shows the rate for the GPU it picked before it starts. This page does not quote a fixed price.

See pricing for how billing works.

Proof

No tested run is recorded for this integration yet. A receipt and example output will be added here after a real run, not before.

Troubleshooting

  • Out of memory

    Use 4-bit loading (QLoRA), a shorter sequence length or a smaller batch size.

  • Adapter not in the output

    Save the adapter into the job's output folder at the end of the script.

  • Gated base model fails

    Accept the licence on the Hub and provide an access token.

Developer reference

Ready to start with Unsloth?

Every run or service has a spending limit you set before it starts.