whereismyai
PyCaret
PyCaret is an open-source AutoML library and a self-hosted platform for running machine-learning experiments.
PyCaret is an open-source machine-learning platform whose engine is a scikit-learn AutoML library published on PyPI. Version 3.4.0 is the frozen 3.x release. Version 4.0, still in progress on the main branch, adds a FastAPI control plane and a React web UI for workspaces, projects, experiments, a model registry, deployments, approvals, monitoring, drift, lineage, webhooks, schedules and LLM advisory copilots. docker compose up starts the UI on localhost:3020 with SQLite in a volume. The same code is meant to split later onto separate API, worker and runtime services. An agent guide is included in the repository.
PyCaret at a glance
- Best for: Software engineer, Researcher
- Runs on: open source, self-hosted, Web
- Pricing: Free and open source
- Released: November 2019
- Category: AI coding tool
- Website: pycaret.org
Key features
- sklearn-based AutoML engine on PyPI (3.4.0)
- FastAPI control plane for experiments and a model registry
- React web UI started with docker compose
- Drift, lineage, webhooks and scheduled runs
- LLM advisory copilots in the 4.0 line
- Single-box local install that can later split into services
Use it for
- Compare classification models with a few lines of Python
- Run a local experiment UI without a cloud MLOps suite
- Keep a model registry and approval flow on one machine
Alternatives to PyCaret
- CopilotKit — Connects your application to AI agents with generative UI and human approvals.
- Runcell — It works directly inside Jupyter notebooks to execute code and read outputs.
- OpenCode — Works with 75+ model providers, local models, and existing subscriptions
- MLflow — Complete observability and evaluation for production AI apps.
- XY — Renders millions of points with native Rust kernels and plain CSS styling.
- Genkit — Provides unified APIs and local developer tools for AI application workflows.
- Evidently — Provides 100+ built-in metrics and test suites for machine learning and LLM systems.
- GitHub Copilot — Built into GitHub, with Claude and Codex agents on the same plan
- CodeGraph — It eliminates file crawling, reducing AI agent tool calls by 88% and cutting costs.
- ClearML — Provides an end-to-end platform for infrastructure control, AI development, and GenAI deployment.
- Jupyter AI — Connects frontier AI agents directly to JupyterLab notebooks.
- Codex Autoresearch — It automates code experiments by managing Git commits and verification loops.
All PyCaret alternatives