From notebook to production

Deploy your model
in minutes.

Train in Studio or Colab, import your model, and get a live API. Query with code, chat, or a spreadsheet.

predict.py
import requests

response = requests.post(
  "https://api.mldash.com/v1/predict/churn",
  headers={"X-API-Key": "mldash_your_api_key"},
  json={
    "tenure_months": 24,
    "monthly_charges": 79.99,
    "support_tickets": 3
  }
)

result = response.json()
print(result)
Response
Waiting for response...

1.2M+

predictions served

48ms

avg response

5

ways to consume

Full ML lifecycle
Multi-feature support
48ms avg response

Everything you need

From training to sharing, one platform handles it all.

Studio

Write Python in the browser, train any model, and export to ML-Dash.

SDK Export

Bring your own model from Colab or Jupyter. Export with mldash.export().

Instant API

Every model gets a REST endpoint automatically. Copy your key and go.

Chat Interface

Non-technical users query models in plain English via chat.

Explore & Share

Publish models to the community. Anyone can discover, try, and bookmark them.

Multi-Feature

Combine text, numerical, and categorical features in a single model.

Three steps to production

Go from a trained model to a live API in under 5 minutes.

1

Create

Train in Studio (our in-browser notebook) or in your own environment — Colab, Jupyter, VS Code.

2

Deploy

Import your model or export from Studio. Your REST endpoint is ready instantly with API key auth.

3

Share

Python, cURL, JavaScript, Dash AI chat, or the Explore marketplace. Your model, accessible to everyone.

API Endpoints

Instant API endpoints
for every model

Every model you train gets a REST API endpoint automatically. No infrastructure to manage, no setup scripts to write.

python
import requests

response = requests.post(
    "https://api.mldash.app/api/v1/predict/your-model-slug/",
    headers={"X-API-Key": "mldash_your_api_key"},
    json={
        "text": "This product is amazing, highly recommend!",
        "features": {
            "category": "electronics",
            "price": 29.99
        }
    }
)

result = response.json()
print(f"Prediction: {result['prediction']}")
print(f"Model: {result['model_name']}")

48ms avg latency

Optimized inference pipeline for instant results

API key auth

Secure access with mldash_ prefixed keys and rate limiting

3 integrations

Python, JavaScript, and cURL examples included

Customer Churn PredictorActive

Will this customer churn? They have been with us for 8 months, pay $89.99/month, and have opened 5 support tickets.

Based on the Customer Churn Predictor model, here is the prediction:

High Churn Risk

Confidence: 87%

Here are the key factors driving this prediction:

Support tickets (5)
+32%
Short tenure (8 mo)
+28%
High monthly charges ($89.99)
+15%
Contract type (month-to-month)
+12%
Chat Interface

Talk to your models,
no code needed

Skip the API calls. Describe your data, get a prediction, and ask follow-up questions about what drove the result.

Describe your data conversationally, get structured predictions

Ask why a prediction was made and see feature importance scores

Test different inputs to see how predictions change

Quick Start

From notebook to API

Export from Studio or import from your notebook. Your endpoint is ready.

studio.py
# In ML-Dash Studio (or Colab)
import mldash

# Export your trained model
mldash.export(
  model=model,
  name="my-model",
  target={"label": "categorical"},
)

# Your API endpoint is ready at:
# /api/v1/predict/my-model/
Integrations

Works everywhere
you do

Every trained model gets a REST endpoint. Call it from Python, JavaScript, or cURL.

PythonJavaScriptcURL
python
import requests

response = requests.post(
    "https://api.mldash.app/api/v1/predict/your-model-slug/",
    headers={"X-API-Key": "mldash_your_api_key"},
    json={
        "text": "This product is amazing, highly recommend!",
        "features": {
            "category": "electronics",
            "price": 29.99
        }
    }
)

result = response.json()
print(f"Prediction: {result['prediction']}")
print(f"Model: {result['model_name']}")
Model Gallery

Explore community models

Browse models built by the community. Try before you integrate.

Customer Churn Predictor

ML-Dash Team

classification

Predicts customer churn based on usage patterns, support tickets, and billing history.

45.2K
Try it

Sentiment Analyzer

ML-Dash Team

classification

Classifies text as positive, negative, or neutral with confidence scores.

128.4K
Try it

House Price Predictor

Sarah Chen

regression

Estimates residential property prices using location, size, and market conditions.

23.4K
Try it

Email Spam Classifier

Alex Kim

classification

Identifies spam emails using NLP and behavioral features with high recall.

1.2M
Try it

Credit Risk Scorer

David Lee

classification

Assesses credit risk for loan applications based on financial history.

389.0K
Try it

Customer Segmenter

James Wilson

classification

Segments customers into behavioral clusters for targeted marketing.

34.1K
Try it

Frequently asked questions

Everything you need to know about ML-Dash.

Ready to share
your model?

Upload a trained model, get an API endpoint, and start making predictions. That simple.