mlflow/mlflow
The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while controlling costs and managing access to models and data. observed · 2026-08-28
Health v2 · maintenance only
99/100
- Activity 99
- Release rhythm 99
- Longevity 100
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 12.5
- age_days: 3011
- days_rel: 7
- days_push: 7
- n_releases_24m: 51
Adoption not part of the score
27688 stars · 6227 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded
MLflow is an open source AI engineering platform for agents, LLMs, and traditional ML models, providing experiment tracking, model registry, OpenTelemetry-compatible tracing, evaluation with LLM judges, prompt management, and an AI Gateway. It supports the full lifecycle from prototype to production with 100+ framework integrations and a FastAPI-based agent server for deployment.
Use cases
- track machine learning experiments and compare model runs
- trace and debug LLM applications and AI agents
- evaluate LLM output quality with LLM-as-a-judge metrics
- version and manage prompts with lineage tracking
- deploy AI agents to production endpoints
- monitor production LLM app quality, cost, and safety
- manage model versions and deploy models to production
- route LLM requests through a unified OpenAI-compatible gateway
When to choose
- you need end-to-end MLOps or LLMOps tooling in one platform
- you want vendor-neutral, OpenTelemetry-compatible LLM observability
- you need systematic evaluation and regression detection for AI apps
- you want experiment tracking and a model registry for ML teams
- you use multiple LLM providers and need unified access control and cost management
When to avoid
- you only need lightweight logging without a tracking server or UI
- you need a fully managed SaaS with zero self-hosting effort
- your use case is simple LLM API calls with no evaluation or tracing needs
- you require a non-Python-first SDK as your primary interface
Facets
framework · maturity stable
machine-learning llm-training agent-framework monitoring tracing benchmarking api-gateway prompt-engineering rag data-science developer-tools machine-learning large-language-models data-science developer-tools monitoring python cross-platform self-hosted cloud cli mlops experiment-tracking model-registry llm-evaluation llm-tracing opentelemetry prompt-management ai-gateway model-deployment llm-judges agent-observability apache-2.0 ai-agents docker
10 sources
- readme: https://github.com/mlflow/mlflow · fetched 2026-08-28 · 5c5069c14487
- homepage: https://mlflow.org · fetched 2026-08-29 · eeac466268bf
- site_page: https://mlflow.org/docs/latest/genai · fetched 2026-08-29 · df63a676b581
- site_page: https://mlflow.org/docs/latest/ml · fetched 2026-08-29 · 1ec9d46723ce
- site_page: https://mlflow.org/docs/latest/genai/tracing · fetched 2026-08-29 · c76f8ba36acc
- site_page: https://mlflow.org/docs/latest/genai/eval-monitor · fetched 2026-08-29 · 6eec5b4b801d
- site_page: https://mlflow.org/docs/latest/genai/tracing/prod-tracing · fetched 2026-08-29 · 3c115136a65c
- site_page: https://mlflow.org/docs/latest/genai/tracing/quickstart · fetched 2026-08-29 · cb9ae584a026
- site_page: https://mlflow.org/docs/latest/genai/eval-monitor/ai-insights/detect-issues · fetched 2026-08-29 · f71c155ad62b
- registry_pypi: https://pypi.org/pypi/mlflow/json · fetched 2026-08-29 · b318ec70c87d
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| mlflow/mlflow | main | 99 |
For agents
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem