Skip to main content
Baseten trains models on managed GPUs and deploys the resulting checkpoints to production inference on the same platform. Train directly with Loops, a Python SDK for LoRA fine-tuning and RL on a curated set of base models, or bring your own container and training code with Training Jobs. If you run your own RL trainer, RL Rollouts serves its rollouts from a dedicated deployment that follows each checkpoint.

​
Choose your path

  • Fine-tuning or running RL on a supported base model: Loops provisions a dedicated trainer and paired sampler, and each training step is an API call from a Python loop you write. You bring the data; Baseten runs the infrastructure. Start with the Loops overview.
  • Already training elsewhere: Your Axolotl config, TRL script, or custom loop runs unchanged in a container. Baseten provisions the GPUs, syncs checkpoints as your job saves them, and deploys any checkpoint as a production endpoint. That’s Training Jobs.
  • Running RL with your own trainer: Your trainer publishes checkpoints to your own bucket, and a dedicated rollout deployment swaps in each new policy while it keeps serving. See RL Rollouts.
Training Jobs runs your program to completion on hardware you choose. With Loops, your program calls a trainer and sampler through the API.

​
Next steps

Loops quickstart

Train a step, sample from the tuned weights, and deploy a checkpoint.

Jobs quickstart

Run your first training job and deploy the result.

Deploy RL Rollouts

Set your bucket and push a rollout deployment that follows your trainer.

ML Cookbook

Production-ready examples for frameworks and models.