MLflow (mlflow)

MLflow is an Apache 2.0 open-source platform for the end-to-end ML and GenAI lifecycle: tracking, model registry, deployment, evaluation, traces, prompts, and GenAI gateway. The tracking server exposes a REST API under /api/2.0/mlflow.

URL: Visit APIs.json URL

Run: Capabilities Using Naftiko

Type

• x-type: opensource

APIs

• MLflow REST API - /api/2.0/mlflow on the tracking server. Experiments, runs, params, metrics, tags, registered models, model versions, model aliases, artifacts, traces, prompts, datasets, AI gateway. JSON over HTTP.
• MLflow AI Gateway API - Unified HTTP interface to LLM providers with routing, rate-limiting, secrets.

Tags

• ML, MLOps, GenAI, Experiment Tracking, Open Source

Timestamps

• Created: 2026-05-08
• Modified: 2026-05-08

Common Properties

• Website
• Documentation
• Source (Apache 2.0)
• Commercial: Databricks Managed MLflow
• Plans
• RateLimits
• FinOps

Notes

• FOSS, no commercial API surface. Self-host the tracking server. Databricks Managed MLflow is the canonical hosted offering; DagsHub also exposes a per-repo MLflow endpoint.
• REST API is well-documented at https://mlflow.org/docs/latest/rest-api.html.
• Authentication is deployment-specific (basic/OIDC/none).

Maintainers

FN: Kin Lane

Email: kin@apievangelist.com