RunDPO Beta

Train your models to follow instructions using Direct Preference Optimization

Get Started in Seconds

pip install rundpo

Quickstart Tutorial Go to Platform Documentation

Simple API

Train your models with just a few lines of code, without worrying about infrastructure or GPU management.

Powerful Platform

Monitor your training progress, manage API keys, and download your models through our web interface.

Transparent Pricing

Simple, usage-based pricing at $3.28/GPU-hour, with no hidden fees.

Train Your Model

from rundpo import RundpoClient

# Initialize the client
client = RundpoClient()

# Upload your data file
file_upload = client.upload_file("dpo_train.jsonl")

# Start DPO training
run_id = client.run_dpo({
    "file_id": file_upload.file_id,
    "base_model": "meta-llama/Meta-Llama-3.1-8B-Instruct",
    "gpus": 2
})