#126126th of 126 in ML frameworks and libraries
aim
Self-hosted ML experiment tracker with a web UI and Python query API
- Stars
- 6.3k
- License
- Apache-2.0
- Last commit
- Dec 2025
- Last release
- May 2025
- Language
- Python
6,021 on 4 Apr 20266,290 on 31 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Aim logs training runs and other metadata through a Python SDK. Its web UI groups, aggregates and compares runs, and queries are written as Python expressions. The README says it is built to handle tens of thousands of runs, and the SDK gives programmatic access to tracked data, for example from Jupyter notebooks.
Who it is for: ML engineers who want a self-hosted tracker for training runs
Strengths
- Query runs with Python expressions in the UI and through the SDK
- Public live demos on play.aimstack.io with real training logs
- Tracks system info and resource usage; supports alerting and notifications
- Integrates with popular ML frameworks and offers migration from other trackers
Weaknesses
- README lists Linux and macOS only; no Windows support stated
- Last release 2025-05-08, while the latest commit is 2025-12-31
- Enterprise support is sold separately; README does not describe multi-user auth
- Focused on training experiments; LLM tracing and prompt management are not described
What it needs
- no GPU
- Docker
Quick answers
Is aim open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run aim with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does aim need a GPU?
No, it runs without one.
Also in ML frameworks and libraries
See all 126| Rank | Project | Score |
|---|---|---|
| 1 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 95 out of 100 |
| 2 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |