whereismyai
Aim
Aim is an open-source, self-hosted tracker for machine-learning experiments and other training metadata.
Aim is an open-source experiment tracker that logs training runs and other AI metadata, then lets people compare them in a self-hosted UI or query them through an SDK. It is built to hold tens of thousands of runs, with grouping, aggregation and Python-expression queries. It records system info and resource use, can alert on training progress, and integrates with popular ML frameworks, with a documented path to migrate from other trackers. Jupyter notebooks and automations use the same metadata API. AimStack sells enterprise support on top of the core project. Documentation and examples live beside the GitHub repository.
Aim at a glance
- Best for: Software engineer, Researcher
- Runs on: open source, self-hosted, Web
- Pricing: Free and open source; AimStack offers paid enterprise support
- Released: May 2019
- Category: AI coding tool
- Website: aimstack.io
Key features
- Self-hosted UI to compare training runs
- SDK and Python-expression queries
- Grouping, aggregation and resource tracking
- Integrations with popular ML frameworks
- Migration path from other experiment trackers
- Optional enterprise support from AimStack
Use it for
- Compare hyperparameters across many training runs
- Query logged metrics from a Jupyter notebook
- Replace a hosted tracker with a local install
Alternatives to Aim
- Evidently — Provides 100+ built-in metrics and test suites for machine learning and LLM systems.
- SwanLab — Tracks training metrics and system performance with minimal code integration.
- aimAndShoot — Bots evolve continuously through shared player data and genetic algorithms.
- Jupyter AI — Connects frontier AI agents directly to JupyterLab notebooks.
- openGym — Zero subscriptions, accounts, or telemetry, with support for bring-your-own AI coaching.
- Genkit — Provides unified APIs and local developer tools for AI application workflows.
- pstack — Improves AI coding agent outcomes with rigorous workflows.
- OpenCode — Works with 75+ model providers, local models, and existing subscriptions
- Ray — It orchestrates distributed infrastructure for any AI workload at any scale.
- MLflow — Complete observability and evaluation for production AI apps.
- Kiln — Combines a user-friendly desktop app with an open-source Python library.
- Runcell — It works directly inside Jupyter notebooks to execute code and read outputs.
All Aim alternatives