Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
orchestra-research avatar

Mlflow

orchestra-research/ai-research-skills

  • 521 installs
  • 11.2k repo stars
  • Updated June 16, 2026
  • orchestra-research/ai-research-skills

mlflow is a deployment skill from orchestra-research/ai-research-skills that ships a trained MLflow model from the registry to local REST, Docker, cloud managed endpoints, or batch jobs with a repeatable production deplo

About

mlflow is a production deployment guide skill in orchestra-research/ai-research-skills for ML engineers moving trained models out of the MLflow registry. The skill documents six deployment targets—local server, REST API, Docker, AWS SageMaker, Azure ML, and Kubernetes—plus batch inference and monitoring patterns, each with complexity ratings from low to high. Developers reach for mlflow when a registered model is validated and needs a repeatable path to serving infrastructure, whether containerized Docker deploys, managed cloud endpoints, or batch scoring jobs, with production checklists for REST serving and orchestration.

  • Compares seven deployment targets (local server, REST API, Docker, SageMaker, Azure ML, Kubernetes, batch) with complexi
  • Documents `mlflow models serve` for registry paths and run artifacts, including host, port, and worker flags
  • Includes curl examples for single and batch `/invocations` JSON payloads against a local server
  • Covers REST API serving, Docker images, cloud managed paths, batch offline inference, and production monitoring sections
  • Table-of-contents structure spans deployment options through monitoring for end-to-end ML ops

Mlflow by the numbers

  • 521 all-time installs (skills.sh)
  • +32 installs in the week ending Jul 26, 2026 (Skillselion tracking)
  • Ranked #432 of 2,066 Data Science & ML skills by installs in the Skillselion catalog
  • Data as of Jul 28, 2026 (Skillselion catalog sync)
npx skills add https://github.com/orchestra-research/ai-research-skills --skill mlflow

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs521
repo stars★ 11.2k
Last updatedJune 16, 2026
Repositoryorchestra-research/ai-research-skills ↗

How do you deploy an MLflow model to production?

Ship a trained MLflow model from the registry to local REST, Docker, cloud managed endpoints, or batch jobs with a repeatable deployment checklist.

Who is it for?

ML engineers with a registered MLflow model ready for production serving across REST, Docker, or cloud managed endpoints.

Skip if: Teams still in hyperparameter tuning or cross-validation who have not registered a model in MLflow yet.

When should I use this skill?

A trained model in the MLflow registry needs deployment to REST, Docker, cloud endpoints, Kubernetes, or batch jobs.

What you get

Deployed MLflow model on chosen target with serving endpoint, Docker image, or batch inference job and monitoring hooks.

  • deployed model endpoint
  • Docker serving image
  • batch inference job config

By the numbers

  • Documents 6 deployment targets: local server, REST API, Docker, AWS SageMaker, Azure ML, Kubernetes
  • Includes batch inference as a separate production deployment option

Files

SKILL.mdMarkdownGitHub ↗

MLflow: ML Lifecycle Management Platform

When to Use This Skill

Use MLflow when you need to:

  • Track ML experiments with parameters, metrics, and artifacts
  • Manage model registry with versioning and stage transitions
  • Deploy models to various platforms (local, cloud, serving)
  • Reproduce experiments with project configurations
  • Compare model versions and performance metrics
  • Collaborate on ML projects with team workflows
  • Integrate with any ML framework (framework-agnostic)

Users: 20,000+ organizations | GitHub Stars: 23k+ | License: Apache 2.0

Installation

# Install MLflow
pip install mlflow

# Install with extras
pip install mlflow[extras]  # Includes SQLAlchemy, boto3, etc.

# Start MLflow UI
mlflow ui

# Access at http://localhost:5000

Quick Start

Basic Tracking

import mlflow

# Start a run
with mlflow.start_run():
    # Log parameters
    mlflow.log_param("learning_rate", 0.001)
    mlflow.log_param("batch_size", 32)

    # Your training code
    model = train_model()

    # Log metrics
    mlflow.log_metric("train_loss", 0.15)
    mlflow.log_metric("val_accuracy", 0.92)

    # Log model
    mlflow.sklearn.log_model(model, "model")

Autologging (Automatic Tracking)

import mlflow
from sklearn.ensemble import RandomForestClassifier

# Enable autologging
mlflow.autolog()

# Train (automatically logged)
model = RandomForestClassifier(n_estimators=100, max_depth=5)
model.fit(X_train, y_train)

# Metrics, parameters, and model logged automatically!

Core Concepts

1. Experiments and Runs

Experiment: Logical container for related runs Run: Single execution of ML code (parameters, metrics, artifacts)

import mlflow

# Create/set experiment
mlflow.set_experiment("my-experiment")

# Start a run
with mlflow.start_run(run_name="baseline-model"):
    # Log params
    mlflow.log_param("model", "ResNet50")
    mlflow.log_param("epochs", 10)

    # Train
    model = train()

    # Log metrics
    mlflow.log_metric("accuracy", 0.95)

    # Log model
    mlflow.pytorch.log_model(model, "model")

# Run ID is automatically generated
print(f"Run ID: {mlflow.active_run().info.run_id}")

2. Logging Parameters

with mlflow.start_run():
    # Single parameter
    mlflow.log_param("learning_rate", 0.001)

    # Multiple parameters
    mlflow.log_params({
        "batch_size": 32,
        "epochs": 50,
        "optimizer": "Adam",
        "dropout": 0.2
    })

    # Nested parameters (as dict)
    config = {
        "model": {
            "architecture": "ResNet50",
            "pretrained": True
        },
        "training": {
            "lr": 0.001,
            "weight_decay": 1e-4
        }
    }

    # Log as JSON string or individual params
    for key, value in config.items():
        mlflow.log_param(key, str(value))

3. Logging Metrics

with mlflow.start_run():
    # Training loop
    for epoch in range(NUM_EPOCHS):
        train_loss = train_epoch()
        val_loss = validate()

        # Log metrics at each step
        mlflow.log_metric("train_loss", train_loss, step=epoch)
        mlflow.log_metric("val_loss", val_loss, step=epoch)

        # Log multiple metrics
        mlflow.log_metrics({
            "train_accuracy": train_acc,
            "val_accuracy": val_acc
        }, step=epoch)

    # Log final metrics (no step)
    mlflow.log_metric("final_accuracy", final_acc)

4. Logging Artifacts

with mlflow.start_run():
    # Log file
    model.save('model.pkl')
    mlflow.log_artifact('model.pkl')

    # Log directory
    os.makedirs('plots', exist_ok=True)
    plt.savefig('plots/loss_curve.png')
    mlflow.log_artifacts('plots')

    # Log text
    with open('config.txt', 'w') as f:
        f.write(str(config))
    mlflow.log_artifact('config.txt')

    # Log dict as JSON
    mlflow.log_dict({'config': config}, 'config.json')

5. Logging Models

# PyTorch
import mlflow.pytorch

with mlflow.start_run():
    model = train_pytorch_model()
    mlflow.pytorch.log_model(model, "model")

# Scikit-learn
import mlflow.sklearn

with mlflow.start_run():
    model = train_sklearn_model()
    mlflow.sklearn.log_model(model, "model")

# Keras/TensorFlow
import mlflow.keras

with mlflow.start_run():
    model = train_keras_model()
    mlflow.keras.log_model(model, "model")

# HuggingFace Transformers
import mlflow.transformers

with mlflow.start_run():
    mlflow.transformers.log_model(
        transformers_model={
            "model": model,
            "tokenizer": tokenizer
        },
        artifact_path="model"
    )

Autologging

Automatically log metrics, parameters, and models for popular frameworks.

Enable Autologging

import mlflow

# Enable for all supported frameworks
mlflow.autolog()

# Or enable for specific framework
mlflow.sklearn.autolog()
mlflow.pytorch.autolog()
mlflow.keras.autolog()
mlflow.xgboost.autolog()

Autologging with Scikit-learn

import mlflow
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split

# Enable autologging
mlflow.sklearn.autolog()

# Split data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Train (automatically logs params, metrics, model)
with mlflow.start_run():
    model = RandomForestClassifier(n_estimators=100, max_depth=5, random_state=42)
    model.fit(X_train, y_train)

    # Metrics like accuracy, f1_score logged automatically
    # Model logged automatically
    # Training duration logged

Autologging with PyTorch Lightning

import mlflow
import pytorch_lightning as pl

# Enable autologging
mlflow.pytorch.autolog()

# Train
with mlflow.start_run():
    trainer = pl.Trainer(max_epochs=10)
    trainer.fit(model, datamodule=dm)

    # Hyperparameters logged
    # Training metrics logged
    # Best model checkpoint logged

Model Registry

Manage model lifecycle with versioning and stage transitions.

Register Model

import mlflow

# Log and register model
with mlflow.start_run():
    model = train_model()

    # Log model
    mlflow.sklearn.log_model(
        model,
        "model",
        registered_model_name="my-classifier"  # Register immediately
    )

# Or register later
run_id = "abc123"
model_uri = f"runs:/{run_id}/model"
mlflow.register_model(model_uri, "my-classifier")

Model Stages

Transition models between stages: None → Staging → Production → Archived

from mlflow.tracking import MlflowClient

client = MlflowClient()

# Promote to staging
client.transition_model_version_stage(
    name="my-classifier",
    version=3,
    stage="Staging"
)

# Promote to production
client.transition_model_version_stage(
    name="my-classifier",
    version=3,
    stage="Production",
    archive_existing_versions=True  # Archive old production versions
)

# Archive model
client.transition_model_version_stage(
    name="my-classifier",
    version=2,
    stage="Archived"
)

Load Model from Registry

import mlflow.pyfunc

# Load latest production model
model = mlflow.pyfunc.load_model("models:/my-classifier/Production")

# Load specific version
model = mlflow.pyfunc.load_model("models:/my-classifier/3")

# Load from staging
model = mlflow.pyfunc.load_model("models:/my-classifier/Staging")

# Use model
predictions = model.predict(X_test)

Model Versioning

client = MlflowClient()

# List all versions
versions = client.search_model_versions("name='my-classifier'")

for v in versions:
    print(f"Version {v.version}: {v.current_stage}")

# Get latest version by stage
latest_prod = client.get_latest_versions("my-classifier", stages=["Production"])
latest_staging = client.get_latest_versions("my-classifier", stages=["Staging"])

# Get model version details
version_info = client.get_model_version(name="my-classifier", version="3")
print(f"Run ID: {version_info.run_id}")
print(f"Stage: {version_info.current_stage}")
print(f"Tags: {version_info.tags}")

Model Annotations

client = MlflowClient()

# Add description
client.update_model_version(
    name="my-classifier",
    version="3",
    description="ResNet50 classifier trained on 1M images with 95% accuracy"
)

# Add tags
client.set_model_version_tag(
    name="my-classifier",
    version="3",
    key="validation_status",
    value="approved"
)

client.set_model_version_tag(
    name="my-classifier",
    version="3",
    key="deployed_date",
    value="2025-01-15"
)

Searching Runs

Find runs programmatically.

from mlflow.tracking import MlflowClient

client = MlflowClient()

# Search all runs in experiment
experiment_id = client.get_experiment_by_name("my-experiment").experiment_id
runs = client.search_runs(
    experiment_ids=[experiment_id],
    filter_string="metrics.accuracy > 0.9",
    order_by=["metrics.accuracy DESC"],
    max_results=10
)

for run in runs:
    print(f"Run ID: {run.info.run_id}")
    print(f"Accuracy: {run.data.metrics['accuracy']}")
    print(f"Params: {run.data.params}")

# Search with complex filters
runs = client.search_runs(
    experiment_ids=[experiment_id],
    filter_string="""
        metrics.accuracy > 0.9 AND
        params.model = 'ResNet50' AND
        tags.dataset = 'ImageNet'
    """,
    order_by=["metrics.f1_score DESC"]
)

Integration Examples

PyTorch

import mlflow
import torch
import torch.nn as nn

# Enable autologging
mlflow.pytorch.autolog()

with mlflow.start_run():
    # Log config
    config = {
        "lr": 0.001,
        "epochs": 10,
        "batch_size": 32
    }
    mlflow.log_params(config)

    # Train
    model = create_model()
    optimizer = torch.optim.Adam(model.parameters(), lr=config["lr"])

    for epoch in range(config["epochs"]):
        train_loss = train_epoch(model, optimizer, train_loader)
        val_loss, val_acc = validate(model, val_loader)

        # Log metrics
        mlflow.log_metrics({
            "train_loss": train_loss,
            "val_loss": val_loss,
            "val_accuracy": val_acc
        }, step=epoch)

    # Log model
    mlflow.pytorch.log_model(model, "model")

HuggingFace Transformers

import mlflow
from transformers import Trainer, TrainingArguments

# Enable autologging
mlflow.transformers.autolog()

training_args = TrainingArguments(
    output_dir="./results",
    num_train_epochs=3,
    per_device_train_batch_size=16,
    evaluation_strategy="epoch",
    save_strategy="epoch",
    load_best_model_at_end=True
)

# Start MLflow run
with mlflow.start_run():
    trainer = Trainer(
        model=model,
        args=training_args,
        train_dataset=train_dataset,
        eval_dataset=eval_dataset
    )

    # Train (automatically logged)
    trainer.train()

    # Log final model to registry
    mlflow.transformers.log_model(
        transformers_model={
            "model": trainer.model,
            "tokenizer": tokenizer
        },
        artifact_path="model",
        registered_model_name="hf-classifier"
    )

XGBoost

import mlflow
import xgboost as xgb

# Enable autologging
mlflow.xgboost.autolog()

with mlflow.start_run():
    dtrain = xgb.DMatrix(X_train, label=y_train)
    dval = xgb.DMatrix(X_val, label=y_val)

    params = {
        'max_depth': 6,
        'learning_rate': 0.1,
        'objective': 'binary:logistic',
        'eval_metric': ['logloss', 'auc']
    }

    # Train (automatically logged)
    model = xgb.train(
        params,
        dtrain,
        num_boost_round=100,
        evals=[(dtrain, 'train'), (dval, 'val')],
        early_stopping_rounds=10
    )

    # Model and metrics logged automatically

Best Practices

1. Organize with Experiments

# ✅ Good: Separate experiments for different tasks
mlflow.set_experiment("sentiment-analysis")
mlflow.set_experiment("image-classification")
mlflow.set_experiment("recommendation-system")

# ❌ Bad: Everything in one experiment
mlflow.set_experiment("all-models")

2. Use Descriptive Run Names

# ✅ Good: Descriptive names
with mlflow.start_run(run_name="resnet50-imagenet-lr0.001-bs32"):
    train()

# ❌ Bad: No name (auto-generated UUID)
with mlflow.start_run():
    train()

3. Log Comprehensive Metadata

with mlflow.start_run():
    # Log hyperparameters
    mlflow.log_params({
        "learning_rate": 0.001,
        "batch_size": 32,
        "epochs": 50
    })

    # Log system info
    mlflow.set_tags({
        "dataset": "ImageNet",
        "framework": "PyTorch 2.0",
        "gpu": "A100",
        "git_commit": get_git_commit()
    })

    # Log data info
    mlflow.log_param("train_samples", len(train_dataset))
    mlflow.log_param("val_samples", len(val_dataset))

4. Track Model Lineage

# Link runs to understand lineage
with mlflow.start_run(run_name="preprocessing"):
    data = preprocess()
    mlflow.log_artifact("data.csv")
    preprocessing_run_id = mlflow.active_run().info.run_id

with mlflow.start_run(run_name="training"):
    # Reference parent run
    mlflow.set_tag("preprocessing_run_id", preprocessing_run_id)
    model = train(data)

5. Use Model Registry for Deployment

# ✅ Good: Use registry for production
model_uri = "models:/my-classifier/Production"
model = mlflow.pyfunc.load_model(model_uri)

# ❌ Bad: Hard-code run IDs
model_uri = "runs:/abc123/model"
model = mlflow.pyfunc.load_model(model_uri)

Deployment

Serve Model Locally

# Serve registered model
mlflow models serve -m "models:/my-classifier/Production" -p 5001

# Serve from run
mlflow models serve -m "runs:/<RUN_ID>/model" -p 5001

# Test endpoint
curl http://127.0.0.1:5001/invocations -H 'Content-Type: application/json' -d '{
  "inputs": [[1.0, 2.0, 3.0, 4.0]]
}'

Deploy to Cloud

# Deploy to AWS SageMaker
mlflow sagemaker deploy -m "models:/my-classifier/Production" --region-name us-west-2

# Deploy to Azure ML
mlflow azureml deploy -m "models:/my-classifier/Production"

Configuration

Tracking Server

# Start tracking server with backend store
mlflow server \
  --backend-store-uri postgresql://user:password@localhost/mlflow \
  --default-artifact-root s3://my-bucket/mlflow \
  --host 0.0.0.0 \
  --port 5000

Client Configuration

import mlflow

# Set tracking URI
mlflow.set_tracking_uri("http://localhost:5000")

# Or use environment variable
# export MLFLOW_TRACKING_URI=http://localhost:5000

Resources

  • Documentation: https://mlflow.org/docs/latest
  • GitHub: https://github.com/mlflow/mlflow (23k+ stars)
  • Examples: https://github.com/mlflow/mlflow/tree/master/examples
  • Community: https://mlflow.org/community

See Also

  • references/tracking.md - Comprehensive tracking guide
  • references/model-registry.md - Model lifecycle management
  • references/deployment.md - Production deployment patterns

Related skills

microsoftMicrosoft FoundryDeploy, evaluate, and continuously improve Microsoft Foundry agents from a single agent interface.478k1.3klllllllamaAi Research ReproductionOrchestrate trustworthy, auditable reproduction of deep learning repositories directly from their READMEs.164k507lllllllamaRun TrainSafely execute selected deep learning training commands with standardized evidence capture.164k507lllllllamaExplore RunSafely run isolated exploratory experiments with clear recording and conservative selection before committing changes.164k507lllllllamaPaper Context ResolverFetch precise reproduction-critical details like dataset splits, preprocessing steps, or evaluation protocols from the original academic paper when the repo README leav141k507lllllllamaRepo Intake And PlanScan unfamiliar AI research repositories and receive a minimal, trustworthy reproduction target before investing significant time.140k507

How it compares

Pick mlflow when deploying from the MLflow model registry and you need target-specific checklists across REST, containers, and managed cloud endpoints.

FAQ

Which deployment targets does the MLflow skill cover?

The mlflow skill from orchestra-research/ai-research-skills covers local server, REST API, Docker, AWS SageMaker, Azure ML, Kubernetes, and batch inference. Each target includes complexity guidance from low to high.

Does the MLflow skill include production monitoring?

The mlflow deployment guide includes production patterns and monitoring alongside serving options. The skill walks from MLflow registry export through REST, containerized, cloud managed, and batch deployment checklists.

Data Science & MLllmautomation
Browse by category
AI & AgentsFrontend & DesignBackend & DataDev ToolsAutomationTesting & ReviewSecurityMonitoring & CloudDocs & Planning
Design & frontend
Design systemsAnimation & motionAccessibility (a11y)CSS & TailwindUI designUXDesign to codeshadcn/uiLanding pagesBest design skills
Most searched skills
find-skillscavemanfrontend-designbrainstormingsystematic-debuggingstop-slopemil-design-engui-ux-pro-maxkarpathy-guidelinesweb-design-guidelinesreact-best-practicesgrill-meimprovemicrosoft-foundryagent-browser
Browse all listings
All agent skillsAll MCP serversAll marketplacesAll plugins

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.