Job Launchers#

NeMo AutoModel provides several ways to launch training. The right choice depends on your hardware and environment.

Which Launcher Should I Use?#

Launcher	Best for	GPUs	Guide
Local Workstation	Getting started, debugging, single-node training	1-8 on one machine	Local Workstation
Slurm	Multi-node batch jobs on HPC clusters	8+ across nodes	Slurm
NeMo-Run	Managed execution on Slurm, Kubernetes, Docker, local	1+	NeMo-Run
SkyPilot	Cloud training (AWS, GCP, Azure) with spot pricing	Any	SkyPilot

I have 1-2 GPUs on my workstation#

Use the interactive launcher. No scheduler or cluster software needed:

automodel examples/llm_finetune/llama3_2/llama3_2_1b_squad.yaml

See the Local Workstation guide.

I have access to a Slurm cluster#

Add a slurm: section to your YAML config and submit with the same automodel command. The CLI generates the torchrun invocation and calls sbatch for you:

automodel config_with_slurm.yaml

See the Slurm guide.

I want managed job submission (Slurm, Kubernetes, Docker)#

Add a nemo_run: section to your YAML config. NeMo-Run loads a pre-configured executor for your compute target and submits the job:

automodel config_with_nemo_run.yaml

See the NeMo-Run guide.

I want to train on the cloud#

Add a skypilot: section to your YAML config. SkyPilot provisions VMs on any major cloud and handles spot-instance preemption automatically:

automodel config_with_skypilot.yaml

See the SkyPilot guide.

All Launchers Use the Same Config#

Every launcher shares the same YAML recipe format. The only difference is an optional launcher section (slurm:, nemo_run:, or skypilot:) that tells the CLI where to run. Without a launcher section, training runs interactively on the current machine.