Experiment tracking
MLflow
Open-source platform for tracking machine learning experiments (parameters, metrics, artifacts), versioning models in a model registry, and tracing and evaluating LLM and agent applications. Teams run its tracking server themselves or use a managed MLflow service.
- Pricing model
- Free and open source
- Free tier
- Yes
- Self-hostable
- Yes
- Open source
- Yes (Apache-2.0)
MLflow works with 13 of the LLM providers we track
Each is documented in MLflow's own docs, shown with its lowest current rate from the prices we track, per 1M input tokens. MLflow may support providers we don't track, and can reach any OpenAI-compatible endpoint; its documentation (opens in a new tab) has the full list.
