Configs2026年4月1日·1 分钟阅读

MLflow — Open Source AI Engineering Platform

MLflow is the largest open-source AI engineering platform for tracing, evaluation, prompt management, and model deployment. 25K+ GitHub stars. 60M+ monthly downloads. Apache 2.0.

TO
TokRepo精选 · Community
快速使用

先拿来用，再决定要不要深挖

这里应该同时让用户和 Agent 知道第一步该复制什么、安装什么、落到哪里。

# Install and start server
pip install mlflow
mlflow server

# Or with uvx (no install)
uvx mlflow server

# Track an experiment
python -c "
import mlflow
mlflow.set_tracking_uri('http://localhost:5000')
with mlflow.start_run():
    mlflow.log_param('model', 'gpt-4o')
    mlflow.log_metric('accuracy', 0.95)
    mlflow.log_metric('latency_ms', 120)
    print('Logged to MLflow!')
"
# View at http://localhost:5000
介绍

MLflow is the largest open-source platform for AI engineering, enabling teams to debug, evaluate, monitor, and optimize production AI applications for agents, LLMs, and ML models. With 25,000+ GitHub stars, 60+ million monthly downloads, and Apache 2.0 license, MLflow provides observability and tracing (OpenTelemetry-based), 50+ built-in evaluation metrics, prompt management with version control, an AI Gateway for unified multi-provider API access with rate limiting and cost control, experiment tracking, model registry, and deployment tools. Supports 60+ frameworks including LangChain, OpenAI, and Anthropic.

Best for: ML/AI teams needing end-to-end experiment tracking, model management, and production monitoring Works with: Claude Code, OpenAI Codex, Cursor, Gemini CLI, Windsurf Integrations: 60+ frameworks (LangChain, OpenAI, Anthropic, PyTorch, etc.)

Key Features

  • Observability: OpenTelemetry-based tracing for LLM applications
  • 50+ evaluation metrics: Built-in LLM judges for quality assessment
  • Prompt management: Version, test, deploy prompts with lineage
  • AI Gateway: Unified API across providers with rate limiting and cost control
  • Model registry: Version, stage, and deploy models
  • 60+ integrations: LangChain, OpenAI, Anthropic, PyTorch, and more
  • Multi-language: Python, TypeScript, Java, OpenTelemetry

FAQ

Q: What is MLflow? A: MLflow is the largest open-source AI engineering platform with 25K+ stars. Tracing, 50+ eval metrics, prompt management, AI Gateway, model registry. 60M+ monthly downloads. Apache 2.0.

Q: How do I install MLflow? A: pip install mlflow then mlflow server. Or uvx mlflow server without installing. Access UI at localhost:5000.

🙏

来源与感谢

Created by Databricks. Licensed under Apache 2.0. mlflow/mlflow — 25,000+ GitHub stars

相关资产

Void — Open-Source Cursor Alternative

Void is an open-source AI code editor alternative to Cursor. 28.5K+ stars. Checkpoints, custom models, local hosting, built on VS Code. Apache 2.0.

TokRepo精选

AI Shell — Natural Language to Shell Commands

AI Shell converts natural language into shell commands using AI. 5.2K+ stars. 13+ languages, chat mode, custom endpoints. MIT. By Builder.io.

TokRepo精选

LLM — CLI Tool for 100+ Language Models

LLM is a CLI and Python library for accessing 100+ LLMs via APIs or locally. 11.5K+ stars. SQLite logging, embeddings, structured data. Apache 2.0.

TokRepo精选
◈首页🏆Trending👤我的