About Neptune
Neptune is the most scalable experiment tracker for teams that train foundation models. It brings scalability and features that make it easier to monitor and debug large-scale training.
note
Neptune is currently available only to select customers. Get access to Neptune
API
The Neptune API consists of two Python packages:
- neptune-scale, for logging metadata to Neptune.
- neptune-fetcher, for fetching logged metadata from Neptune.
Web app
The Neptune web app is where you can visualize, compare, and monitor your logged metadata:
- Runs table – Create powerful filters and customize columns. Share views ad-hoc, or save them for everyone.
- Experiments – Monitor long model-training experiments with multiple steps and forks.
- Charts – Visualize and compare thousands of metrics in seconds, with many additional configuration options.
- Side-by-side – Contrast a large number of configs, scores, and other values at once.
- Dashboards – Compose a set of custom widgets for inspecting currently visualized runs.
- Reports – Analyze groups of runs seamlessly, even from different projects. Add text and work on drafts before publishing.
Next steps
- Learn the Neptune concepts
- Follow Get started to install and configure Neptune