> See the full index of documentation pages at https://docs.zettabyte.com/llms.txt.

# Run on GPUs

> Launch GPU compute, queue for capacity, develop in notebooks, run training and fine-tuning jobs, reuse templates, and build container images.

Source: https://docs.zettabyte.com/docs/run-on-gpus

Last updated: 2026-07-21

## Pages

- [GPU instances](https://docs.zettabyte.com/docs/run-on-gpus/gpu-instances.md): Reserve GPU capacity to run notebooks, training, and inference.
- [The resource queue](https://docs.zettabyte.com/docs/run-on-gpus/resource-queue.md): When the capacity you request is not immediately available, you can queue for it instead of failing.
- [Notebooks and IDEs](https://docs.zettabyte.com/docs/run-on-gpus/notebooks-and-ides.md): Open a browser-based `Jupyter` or `VSCode` environment, with or without a GPU.
- [Tasks overview](https://docs.zettabyte.com/docs/run-on-gpus/training-and-tasks.md): Training, fine-tuning, and inference all run as tasks. Start here, then follow the walkthrough for what you want to do.
- [Run a training job](https://docs.zettabyte.com/docs/run-on-gpus/run-a-training-job.md): Run a training script on GPUs, from image to saved weights.
- [Fine-tune a model](https://docs.zettabyte.com/docs/run-on-gpus/fine-tune-a-model.md): Adapt an existing model to your data with the guided fine-tuning flow.
- [Task templates](https://docs.zettabyte.com/docs/run-on-gpus/task-templates.md): Save a task configuration once and reuse it.
- [Container images](https://docs.zettabyte.com/docs/run-on-gpus/container-images.md): Get your code and environment onto GPUs with container images.
- [zCLOUD CLI](https://docs.zettabyte.com/docs/run-on-gpus/zcloud-cli.md): Manage instances, tasks, and files from your terminal.

## Optional

- [Full Run on GPUs documentation](https://docs.zettabyte.com/docs/run-on-gpus/llms-full.txt): every Run on GPUs page as one markdown file