#9696th of 105 in ML frameworks and libraries
pycaret
Self-hosted AutoML platform with experiment tracking, model registry and deployment UI
- Stars
- 9.9k
- License
- custom license
- Last commit
- May 2026
- Last release
- Apr 2024
- Language
- Python
9,667 on 1 Apr 20269,857 on 21 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
PyCaret 4.0 combines a scikit-learn-based AutoML engine with a FastAPI control plane and a React web UI. You upload a dataset, run an experiment that trains about 12 algorithms, promote the winner to a model registry and deploy it behind a prediction endpoint. The README marks 4.0 as work in progress; the default setup runs in-process with SQLite.
Who it is for: Teams self-hosting tabular AutoML with a registry and prediction endpoints
Strengths
- Covers train, register, deploy and predict in one UI and API
- Two-container docker compose setup, web UI on port 3020
- Optional Postgres and S3 storage via environment variables
- Engine is MIT; supports classification, regression, clustering, anomaly and time series
Weaknesses
- 4.0 is work in progress; most Protocol extractions for splitting services are unfinished
- Default setup runs scheduler and compute in the API process with SQLite
- Control plane is BUSL-1.1, not OSI open source; only the engine is MIT
- Last release is 2024-04-28; the 3.x line on PyPI is frozen
What it needs
- RAM ≥ 6 GB
- no GPU
- Compose
- Needs Docker Compose, SQLite (default), Postgres (optional), S3 (optional)
- Models: scikit-learn algorithms, Anthropic API (optional copilots), OpenAI API (optional copilots)
- port 3020
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 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 | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |