Skip to main content
Training Jobs runs your Axolotl configuration, TRL script, or custom training loop in a container on managed GPUs. Baseten provisions the hardware, syncs saved checkpoints, and can deploy any checkpoint as a production endpoint. To compare Training Jobs with Loops, see Training on Baseten.

​
How Training Jobs works

Baseten stores checkpoints while the job runs and can deploy any of them as a production endpoint. You do not need to download and re-upload weights or configure separate serving infrastructure. The core workflow uses two commands:
From job submission to a served model:
  1. Define your job. Declare compute, container image, runtime, and checkpointing in a Python configuration file.
  2. Submit the job. truss train push packages your code and starts the job on H100 or H200 GPUs, using one or multiple nodes.
  3. Sync checkpoints. Baseten stores each checkpoint saved by the job.
  4. Deploy a checkpoint. truss train deploy_checkpoints turns a synced checkpoint into a production endpoint.

​
Supported frameworks

Training Jobs runs the training frameworks and dependencies that you package in the container. Browse the ML Cookbook for more examples including multi-node training with FSDP and DeepSpeed.

​
Key features

​
Checkpoint management

Checkpoints sync automatically to Baseten storage during training. You can:
  • Deploy any checkpoint as a production endpoint with truss train deploy_checkpoints.
  • Download checkpoints for local evaluation and analysis.
  • Resume from any checkpoint if a job fails or you want to train further.
Learn more about checkpointing.

​
BDN weight and data loading

Load model weights and training data through Baseten Delivery Network (BDN). Mount weights from Hugging Face, S3, GCS, R2, or any HTTPS URL directly into your training container with no download code needed. BDN mirrors weights before compute is provisioned, then caches them for faster mounting on subsequent jobs. See storage and data ingestion for setup details.

​
Persistent caching

Cache models, datasets, and preprocessed data between training jobs. The persistent cache avoids downloading the same files for each run. See the training cache guide for configuration options.

​
Multi-node training

Scale training across multiple GPU nodes with InfiniBand networking. Baseten handles node orchestration, communication setup, and environment variables. You set node_count in your configuration. Learn more about multi-node training.

​
Remote access

Connect to running training containers to debug, inspect state, and iterate without resubmitting. Baseten offers two options:
  • SSH: Connect from any OpenSSH client for terminal sessions and file transfer with scp or sftp.
  • VS Code & Cursor: Connect through VS Code or Cursor Remote Tunnels.
See the Remote access overview to choose between them.

​
Next steps

Jobs quickstart

Run your first training job and deploy the result.

Compare Jobs and Loops

Choose between container-driven Jobs and the Loops training SDK.

ML Cookbook

Browse examples for training frameworks and models.

​
Reference