Discover / LLM Ops & Observability
Aim
by aimhubioPython
Open source experiment tracking tool for comparing and visualizing ML and LLM training runs.
Repositoryactive
Maturity: active because commit 4d ago, latest release v3.29.1. Derived from release and commit history, not a rating.
- Stars
- 6.2k
- Forks
- 402
- Downloads / mo
- 127k
- Last commit
- 2026-08-02
- License
- Apache-2.0
- Open issues
- 466
Market and trust evidence
Edition not yet matchedNo exact skills.sh identity match is available for this repository. Repository adoption and freshness remain visible above; install momentum is not inferred.
Trust analysis is a screening signal, not a security warranty. Read the ranking and trust methodology.
In practice
Written by AI from this repository’s README · low confidenceComparing many training runs and their metadata is impossible when results only exist in scattered logs.
Use it when
Use it when you want a self hosted tracker with a UI to compare runs and an API to query them programmatically.
Not the right pick when
The README is mostly badges and logos, so setup and usage details are only in the external documentation.
Capabilities
- logs training runs and AI metadata
- UI to compare and observe runs
- API to query tracked data programmatically
- integrations with common ML frameworks
- Linux and macOS support
Cost: Free and open source
Install
Derived from the published package name in the repository, not from a model.
Video walkthroughs
How To Make an Aimbot For Any Game [ 2025 ]
How to use Aim to compare 100s of experiments in minutes.
Third-party YouTube uploads matched to this tool by title, channel and repository name on 2026-08-03. Not made, reviewed or endorsed by SkillPilot. View counts and publish months are as of the match date and the month is approximate. Nothing loads from YouTube until you press play.
What the repository ships
Has testsHas docsHas examplesCI configured
Detected from the actual files in the repository root.
Latest release v3.29.1
Published 2025-05-08
What's Changed
- [feat] Constant indexing of in-progress Runs by @alberttorosyan in https://github.com/aimhubio/aim/pull/3310
- [fix] Resolve issue of min/max calculation for single point metrics by @mihran113 in https://github.com/aimhubio/aim/pull/3315
- [fix] Use polling observer to make sure new file modifications are detected by @alberttorosyan in https://github.com/aimhubio/aim/pull/3316
- [feat] Mark stalled runs as finished by @alberttorosyan in https://github.com/aimhubio/aim/pull/3314
- [fix] Aim web ui integration in jupyter/colab by @larissapoghosyan in https://github.com/aimhubio/aim/pull/3319
- [fix] Fallback to union db if index is missing by @alberttorosyan in https://github.com/aimhubio/aim/pull/3317
New Contributors
- @larissapoghosyan made their first contribution in https://github.com/aimhubio/aim/pull/3319
Full Changelog: https://github.com/aimhubio/aim/compare/v3.28.0...v3.29.1
Tags
README
<div align="center">
<table>
<tbody>
<tr>
<td>Drop a star to support Aim ⭐</td>
<td>
<a href="https://community.aimstack.io/">Join Aim discord community</a>
<img width="20px" src="https://user-images.githubusercontent.com/13848158/226759622-063b725d-8b3e-4c75-80c7-11fb04b7adf5.png">
</td>
</tr>
</tbody>
</table>
</div>
<div align="center">
<img src="https://user-images.githubusercontent.com/13848158/225620298-9f9293e9-a138-41fd-bd77-21d53d0490b7.png">
<h3>
An easy-to-use & supercharged open-source experiment tracker
</h3>
Aim logs your training runs and any AI Metadata, enables a beautiful UI to compare, observe them and an API to query them programmatically.
</div>
<br/>
<div align="center">
[Platform Support]()
</div>
<div align="center">
<br/>
<kbd>
<img src="https://user-images.githubusercontent.com/13848158/136374529-af267918-5dc6-4a4e-8ed2-f6333a332f96.gif" />
</kbd>
</div>
</br>
<div align="center">
<sub><strong>SEAMLESSLY INTEGRATES WITH:</strong></sub>
<br/>
<br/>
<img src="https://user-images.githubusercontent.com/97726819/225954732-2b263308-8ed8-4df3-810b-704840328e98.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954727-04eccf0e-51ed-4f2d-8f3b-c9a675ca8e8f.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954728-ca2f498d-51a7-487b-bd69-ffb5f0c2af58.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954689-1076998c-42f4-4e31-ab47-d9f39575fda1.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954739-0231d659-3202-4458-9c35-ba92d1f079b8.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954697-ef2c7091-b089-4b68-8543-80ce7275b683.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954743-dbfe1e98-7b9f-446a-9fe4-ad4fd562f3df.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954736-7c52ab5a-6780-4375-a6f8-b394dae3ad31.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954710-36551a71-b26d-4665-af20-44ee452dd5dd.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954707-4bc078b5-ae6f-4847-bc2c-3f81959accb2.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954725-a4d4c32c-75db-470a-b1da-698982faa23c.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954665-8d844747-a857-41b8-9104-7c27a8bdb81a.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954686-b9c8ec57-d4fc-44e1-a4b8-443db381a00f.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954693-94bc20b4-f51a-4130-8d5f-ddee30d81205.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954674-42fbfdb3-0351-492d-9ea3-1d3ab2b545f5.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954678-25f1b626-2cb1-4e7e-ad83-f7c8ab679c6f.png" height="60" />
<img src="https://user-images.githubusercontent.com/97726819/225954702-d18d2706-dc87-4e09-a678-f010f6d95736.png" height="60" />
</div>
<br/>
<div align="center">
<sub><strong>TRUSTED BY ML TEAMS FROM:</strong></sub>
<br/>
<br/>
<img src="https://user-images.githubusercontent.com/97726819/225952594-a21546f4-987f-42bd-9154-c22b50632b55.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225953224-291fbb6d-e31a-4e36-90ba-92a1dc20bacc.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225953339-4fcc3c99-dda2-4529-ac2a-29c8293be323.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225953084-1d88325a-ad5f-49eb-aa67-f064c4b9f39c.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225952955-397e0f26-f01e-4b25-bd70-9f292a6f77c2.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225952682-a843827b-e870-429d-96d8-3b4fd45471cd.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225952831-1c0e36aa-4120-4635-ab4e-bf1bc685477b.png" height="55" />
<img src="https://user-images.githubusercontent.com/97726819/225953432-4b27653a-aa12-4fbf-897c-01349afa2ad0.png" height="55" />
</div>
<br/>
<p align="center">
AimStack offers enterprise support that's beyond core Aim. Contact via <a href="mailto:hello@aimstack.io">hello@aimstack.io</a> e-mail.
</p>
<h3 align="center">
<a href="#-about"><b>About</b></a> •
<a href="#-demos"><b>Demos</b></a> •
<a href="#-ecosystem"><b>Ecosystem</b></a> •
<a href="#-quick-start"><b>Quick Start</b></a> •
<a href="https://github.com/aimhubio/aim/tree/main/examples"><b>Examples</b></a> •
<a href="https://aimstack.readthedocs.io/en/latest/"><b>Documentation</b></a> •
<a href="#-community"><b>Community</b></a> •
<a href="https://aimstack.io/blog"><b>Blog</b></a>
</h3>
ℹ️ About
Aim is an open-source, self-hosted ML experiment tracking tool designed to handle 10,000s of training runs.
Aim provides a performant and beautiful UI for exploring and comparing training runs.
Additionally, its SDK enables programmatic access to tracked metadata — perfect for automations and Jupyter Notebook analysis.
<p align="center">
<strong>Aim's mission is to democratize AI dev tools 🎯 </strong>
</p>
<div align="center">
<img src="https://user-images.githubusercontent.com/13848158/226426018-f7c11c9b-78d9-4ee4-b292-df28b3e8eaa6.jpg" height="140" />
<img src="https://user-images.githubusercontent.com/13848158/226426005-f7e83923-0f92-44a4-88e4-1735a3d3e119.jpg" height="140" />
<img src="https://user-images.githubusercontent.com/13848158/226426015-4f1122d8-c96a-443f-8698-3db942b1972a.jpg" height="140" />
</div>
</br>
<div align="center">
<table>
<tbody>
<tr>
<th>Log Metadata Across Your ML Pipeline 💾</th>
<th>Visualize & Compare Metadata via UI 📊</th>
</tr>
<tr>
<td>
<ul>
<li>ML experiments and any metadata tracking</li>
<li>Integration with popular ML frameworks</li>
<li>Easy migration from other experiment trackers</li>
</ul>
</td>
<td>
<ul>
<li>Metadata visualization via Aim Explorers</li>
<li>Grouping and aggregation</li>
<li>Querying using Python expressions</li>
</ul>
</td>
</tr>
<tr>
<th>Run ML Trainings Effectively ⚡</th>
<th>Organize Your Experiments 🗂️</th>
</tr>
<tr>
<td>
<ul>
<li>System info and resource usage tracking</li>
<li>Real-time alerting on training progress</li>
<li>Logging and configurable notifications</li>
</ul>
</td>
<td>
<ul>
<li>Detailed run information for easy debugging</li>
<li>Centralized dashboard for holistic view</li>
<li>Runs grouping with tags and experiments</li></ul>
</td>
</tr>
</tbody>
</table>
</div>
🎬 Demos
Check out live Aim demos NOW to see it in action.
| Machine translation experiments | lightweight-GAN experiments|
|:---:|:---:|
| <a href="https://play.aimstack.io/nmt/metrics?grouping=O-JTdCJTIyY29sb3IlMjI6JTVCJTIycnVuLnBhcmFtcy5ocGFyYW1zLm1heF9rJTIyJTVELCUyMnN0cm9rZSUyMjolNUIlNUQsJTIyY2hhcnQlMjI6JTVCJTIybmFtZ
Truncated. Read the full README on GitHub ↗