Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
An Open Source, Self-Hosted Platform For Applied Deep Learning Development
| Date | Stars |
|---|---|
| 2026-07-31 | 288 |
| 2026-08-04 | 287 |
| 2026-08-06 | 287 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
**Build Statuses:**
[](https://jenkins.shehanigans.net/job/atlas/job/master/)
[](https://jenkins.shehanigans.net/job/foundations-uat/20/)
[](https://jenkins.shehanigans.net/job/build-artifacts-atlas/)
[](https://jenkins.shehanigans.net/job/build-installer-atlas/)
<p align="center">
<img src="dessa-square-logo.png">
</p>
---
# Atlas: Self-Hosted Machine Learning Platform
Atlas is a flexible Machine Learning platform that consists of a Python SDK, CLI, GUI & Scheduler to help Machine Learning Engineering teams dramatically reduce the model development time & reduce effort in managing infrastructure.
### Development Status
Atlas has evolved very rapidly and has gone though many iterations in Dessa's history.
The latest version is in BETA.
## Features
<p align="center">
<img width="80%" src="https://static.wixstatic.com/media/29a4f1_ffb0c04ef79843e79dbf2b1fa33a70c4~mv2.png/v1/fill/w_1440,h_1024/Time%20series%20forecast.png">
</p>
**Here are few of the high-level features:**
1. _Self-hosted_: run Atlas on a single node e.g. your latop, or multi-node cluster e.g. on-premise servers or cloud clusters (AWS/GCP/etc.)
2. _Job scheduling_: Collaborate with your team by scheduling and running concurrent ML jobs remotely on your cluster & fully utilize your system resources.
3. _Flexibility_: Multiple GPU jobs? CPU jobs? need to use custom libraries or docker images? No problem - Atlas tries to be unopionated where possible, so you can run jobs how you like.
4. _Experiment managment & tracking_: Tag experiments and easily track hyperparameters, metrics, and artifacts such as images, GIFs, and audio clips in a web-based GUI to track the performance of your models.
5. _Reproducibility_: Every job run is recorded and tracked using a job ID so you can reproduce and share any experiment.
6. _Easy to use SDK_: Atlas's easy to use SDK allows you to run jobs programatically allowing you to do multiple hyperparameter optimization runs programatically
7. _Built in [Tensorboard](https://github.com/tensorflow/tensorboard) integration_: We ❤️ Tensorflow - compare multiple Tensorboard-compaitable job runs directly through the Atlas GUI.
8. _Works well with others_: run any Python code with any frameworks.
# Users guide
## Installation
* [MacOS & Linux Quickstart Guide (~8 mins, recommended)](https://docs.atlas.dessa.com/en/latest/linux-mac-installation/)
* [Windows 10 Guide](https://docs.atlas.dessa.com/en/latest/windows-installation/)
* [AWS Cloud installation](https://docs.atlas.dessa.com/en/latest/atlas-on-aws/)
* [GCP Cloud installation](https://docs.atlas.dessa.com/en/latest/atlas-on-gcp/)
* Multi-node cluster deployment:
* [AWS guide](https://docs.atlas.dessa.com/en/latest/team/aws-installation/)
* [On-prem cluster guide](https://docs.atlas.dessa.com/en/latest/team/on-prem-installation/).
## Documentation
Official documentation for Atlas can be found at https://www.docs.atlas.dessa.com/
All docs are hosted on Read the Docs that track the `docs` folder, please open a pull request here to make changes.
## Community
If you have questions that are not addressed in the [documentaExcerpt of 12,566 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:1830bfc0f75e729c, llm:Description: 'Self-Hosted Machine Learning Platform', 'Python SDK, CLI, GUI & Scheduler to help Machine Learning Engineering teams... managing infrastructure'; topics include model-management, ml, machine-learning, data-science, deep-learning, gpu, ai.
matched fp:1830bfc0f75e729c, llm:Description: 'Self-Hosted Machine Learning Platform', 'Python SDK, CLI, GUI & Scheduler to help Machine Learning Engineering teams... managing infrastructure'; topics include model-management, ml, machine-learning, data-science, deep-learning, gpu, ai.
matched fp:1830bfc0f75e729c, llm:Description: 'Self-Hosted Machine Learning Platform', 'Python SDK, CLI, GUI & Scheduler to help Machine Learning Engineering teams... managing infrastructure'; topics include model-management, ml, machine-learning, data-science, deep-learning, gpu, ai.