Turn your data into tomorrow's answers

Turn your data into tomorrow's answers

The time series forecasting
platform built for data scientists.

The time series forecasting platform built for data scientists.

Access production-ready foundation models for time series forecasting. Zero setup time, instant predictions, and automatic fine-tuning. We handle the infrastructure, you focus on outcomes.
Access production-ready foundation models for time series forecasting. Zero setup time, instant predictions, and automatic fine-tuning. We handle the infrastructure, you focus on outcomes.
Trusted by companies who got tired of guessing their numbers.
Trusted by companies who got tired of
guessing their numbers.

Backed by

Inception

Inception

Our Best Foundation Model

Meet

Mimosa

Meet Mimosa, a foundation model that redefines time series forecasting. Outperforms supervised models out of the box, with automated fine-tuning when you need that extra edge.
Meet Mimosa, a foundation model that redefines time series forecasting. Outperforms supervised models out of the box, with automated fine-tuning when you need that extra edge.
Covariates support
Covariates support

Multi-variate forecasting with dynamic and static covariates support. Zero feature engineering overhead.

Multi-variate forecasting with dynamic and static covariates support. Zero feature engineering overhead.

Auto fine tune
Auto fine tune

Optimize model performance with a single API call. Full visibility into training, zero MLOps overhead.

Optimize model performance with a single API call. Full visibility into training, zero MLOps overhead.

Managed service
Managed service

Your forecasting infrastructure on autopilot. From deployment to scaling, we handle the MLOps heavy lifting.

Your forecasting infrastructure on autopilot. From deployment to scaling, we handle the MLOps heavy lifting.

Forecast with Sulie

From zero to forecasting in minutes

Manage your entire forecasting workflow through our Python SDK. Upload datasets, run forecasts on time series data, and control fine-tuning jobs with a clean, intuitive API.
Manage your entire forecasting workflow through our Python SDK. Upload datasets, run forecasts on time series data, and control fine-tuning jobs with a clean, intuitive API.

1

1

1

Create a free account
Create a free account

Create a free account and get full access to our platform. Extensive trial resources available, no credit card required.

Create a free account and get full access to our platform. Extensive trial resources available, no credit card required.

2

2

2

Set up API keys
Set up API keys

Set up your API key and configure the Python SDK in under a minute. Simple integration with your existing notebooks and ML workflows.

Set up your API key and configure the Python SDK in under a minute. Simple integration with your existing notebooks and ML workflows.

3

3

3

Set up datasets and start forecasting
Set up datasets and start forecasting

Import your datasets and get immediate forecasts using our foundation model. Push performance further with automated fine-tuning when needed.

Import your datasets and get immediate forecasts using our foundation model. Push performance further with automated fine-tuning when needed.

Getting started with Sulie SDK

Getting started with Sulie SDK

from sulie import Sulie

client = Sulie(
api_key=os.environ.get("SULIE_API_KEY")
)

# Prepare your data
df = pd.DataFrame(your_data)

# Upload dataset
dataset = client.upload_dataset(
name="product-purchases-v1",
df=df
)

# Forecast on time-series data forecast = client.forecast(
dataset="product-purchases-v1",
horizon=30, # 30 time steps ahead
frequency="D" # Daily frequency
)

from sulie import Sulie

client = Sulie(
api_key="sulie-****"
)

# Prepare your data
df = pd.DataFrame(your_data)

# Upload dataset
dataset = client.upload_dataset(
name="product-purchases-v1",
df=df
)

# Forecast on time-series data forecast = client.forecast(
dataset="product-purchases-v1",
horizon=30, # 30 time steps ahead
frequency="D" # Daily frequency
)

from sulie import Sulie

client = Sulie(
api_key=os.environ.get("SULIE_API_KEY")
)

# Prepare your data
df = pd.DataFrame(your_data)

# Upload dataset
dataset = client.upload_dataset(
name="product-purchases-v1",
df=df
)

# Forecast on time-series data forecast = client.forecast(
dataset="product-purchases-v1",
horizon=30, # 30 time steps ahead
frequency="D" # Daily frequency
)

FAQ

What are foundation models for time series forecasting?
A foundation model for time series forecasting is a large-scale transformer model pre-trained on diverse time series data that captures universal temporal patterns and relationships. Similar to how BERT and GPT models learn language structure, time series foundation models learn to understand trends, seasonality, and variable interactions across different domains and scales. This pre-training enables the model to adapt to new forecasting tasks with no or minimal fine-tuning, effectively transferring knowledge from its training data to your specific use case
How does zero-shot forecasting compare to supervised models?
Our foundation model delivers superior accuracy without training, outperforming supervised models from day one. Unlike traditional approaches that require extensive historical data and training time, you can start forecasting immediately with just your time series data.
What types of time series data can I work with?
All-in-One users can immediately leverage our foundation model for univariate time series forecasting. For teams needing more advanced capabilities, our Enterprise plan unlocks full multi-variate forecasting with support for both static and dynamic covariates (coming soon to all plans).
Do I need ML expertise to use Sulie?
While Sulie is built for ML teams and data scientists, our Python SDK makes it straightforward to integrate into your workflow. If you're comfortable with pandas DataFrames, you can start forecasting in minutes. Advanced users have full control over model parameters and fine-tuning options.
How does automatic fine-tuning work?
When you need extra performance, our automated fine-tuning pipeline adapts the foundation model to your specific patterns. Simply point to your historical data, and our platform handles the optimization process while maintaining full transparency over hyperparameters.
What about scaling and infrastructure management?
We handle all MLOps complexity. Our managed infrastructure automatically scales with your needs, from notebook experiments to production workloads. No need to worry about deployment, scaling, or maintenance - focus on getting value from your forecasts.
Can I try Sulie before committing?
Yes! Create a free account - no credit card required - and get full access to our platform with extensive trial resources. Test our foundation model's capabilities, explore automated fine-tuning, and integrate with your existing workflows.
What are foundation models for time series forecasting?
A foundation model for time series forecasting is a large-scale transformer model pre-trained on diverse time series data that captures universal temporal patterns and relationships. Similar to how BERT and GPT models learn language structure, time series foundation models learn to understand trends, seasonality, and variable interactions across different domains and scales. This pre-training enables the model to adapt to new forecasting tasks with no or minimal fine-tuning, effectively transferring knowledge from its training data to your specific use case
How does zero-shot forecasting compare to supervised models?
Our foundation model delivers superior accuracy without training, outperforming supervised models from day one. Unlike traditional approaches that require extensive historical data and training time, you can start forecasting immediately with just your time series data.
What types of time series data can I work with?
All-in-One users can immediately leverage our foundation model for univariate time series forecasting. For teams needing more advanced capabilities, our Enterprise plan unlocks full multi-variate forecasting with support for both static and dynamic covariates (coming soon to all plans).
Do I need ML expertise to use Sulie?
While Sulie is built for ML teams and data scientists, our Python SDK makes it straightforward to integrate into your workflow. If you're comfortable with pandas DataFrames, you can start forecasting in minutes. Advanced users have full control over model parameters and fine-tuning options.
How does automatic fine-tuning work?
When you need extra performance, our automated fine-tuning pipeline adapts the foundation model to your specific patterns. Simply point to your historical data, and our platform handles the optimization process while maintaining full transparency over hyperparameters.
What about scaling and infrastructure management?
We handle all MLOps complexity. Our managed infrastructure automatically scales with your needs, from notebook experiments to production workloads. No need to worry about deployment, scaling, or maintenance - focus on getting value from your forecasts.
Can I try Sulie before committing?
Yes! Create a free account - no credit card required - and get full access to our platform with extensive trial resources. Test our foundation model's capabilities, explore automated fine-tuning, and integrate with your existing workflows.
What are foundation models for time series forecasting?
A foundation model for time series forecasting is a large-scale transformer model pre-trained on diverse time series data that captures universal temporal patterns and relationships. Similar to how BERT and GPT models learn language structure, time series foundation models learn to understand trends, seasonality, and variable interactions across different domains and scales. This pre-training enables the model to adapt to new forecasting tasks with no or minimal fine-tuning, effectively transferring knowledge from its training data to your specific use case
How does zero-shot forecasting compare to supervised models?
Our foundation model delivers superior accuracy without training, outperforming supervised models from day one. Unlike traditional approaches that require extensive historical data and training time, you can start forecasting immediately with just your time series data.
What types of time series data can I work with?
All-in-One users can immediately leverage our foundation model for univariate time series forecasting. For teams needing more advanced capabilities, our Enterprise plan unlocks full multi-variate forecasting with support for both static and dynamic covariates (coming soon to all plans).
Do I need ML expertise to use Sulie?
While Sulie is built for ML teams and data scientists, our Python SDK makes it straightforward to integrate into your workflow. If you're comfortable with pandas DataFrames, you can start forecasting in minutes. Advanced users have full control over model parameters and fine-tuning options.
How does automatic fine-tuning work?
When you need extra performance, our automated fine-tuning pipeline adapts the foundation model to your specific patterns. Simply point to your historical data, and our platform handles the optimization process while maintaining full transparency over hyperparameters.
What about scaling and infrastructure management?
We handle all MLOps complexity. Our managed infrastructure automatically scales with your needs, from notebook experiments to production workloads. No need to worry about deployment, scaling, or maintenance - focus on getting value from your forecasts.
Can I try Sulie before committing?
Yes! Create a free account - no credit card required - and get full access to our platform with extensive trial resources. Test our foundation model's capabilities, explore automated fine-tuning, and integrate with your existing workflows.

Want to be the first to know when we introduce new things?

©️2024 Sulie. All rights reserved

©️2024 Sulie. All rights reserved

©️2024 Sulie.

All rights reserved