Daily Agriculture Price Forecasts: Powered by AI

Discover the power of AgFlow’s AI-generated agriculture price forecasts, designed to spot trading signals in opaque Grains, Oilseeds, and Vegoils markets. With 150 algorithmic forward curves generated by our AI technology delivered daily, you’ll have access to crucial insights like never before. Plus, our 12-month forward curves are available before midday CET every regular trading day, so you can stay ahead of the game.


Trusted by Thousands of Agricultural Commodities Professionals

4 Reasons to Harness the Power of AI-Generated
Agriculture Price Forecasts

Forward Curves
Market Signals
Risk Management

Fill the Missing Forward Price Gaps

Quickly Assess the Market of What is in Your Books

Rapidly Identify Trading Signals

Accelerate your MTM, VaR, and P&L calculations

AgFlow’s Agriculture Price Forecasts in Numbers

Complete Forward Curves Daily

Years of Historical


Who Benefits From AgFlow’s Agriculture Price Forecasts?

Rely on a daily market price benchmark, even when the market is opaque, with 150+ complete 12-month forward curves to:

Market Research

Market Research

View if a physical price is outside its usual corridor
Spot trading signals

Risk Management

Risk Management

Ensure your traders’ positions are accurately marked
Get a complete view of the forward markets
Fill Cash Markets' Gaps with AgFlow Agriculture Price Forecasts
Forward Curves

Fill Cash Markets' Gaps

  • +120 Complete forward curves daily
  • Predictions are calculated every day
  • All curves are accessible before Midday CET every day
  • Reconstructed and computed by our proprietary Machine Learning algorithm
  • Based on 7 years of historical data

The AgFlow Method to Generate
Agriculture Price Forecasts

Step 1 - Data Collection

We take all available historical data such as spots, forwards, and futures as input.

Step 2 - Aggregation & Outlier Detection

We aggregate the data and curate them via outlier detection. All this, according to a modified interquartile range (IQR) method, and some data-specific constraints.

Step 3 - Full Forward Curve Reconstruction

We then impute the data to reconstruct full historical forward curves. Reconstruction is done through interpolation and for which we use the most correlated future contract as a proxy.

Step 4 - Machine Learning Model Development

As soon as we have gathered a significant dataset of historical curves, we develop a predictive machine-learning model based on LSTM.

Step 5 - Model Robustness

Since a prediction without a measure of confidence is meaningless, we calculate a 95% confidence interval based on a Montecarlo dropout. This gives our clients an indication of the model’s robustness.

Step 6 - Backtesting & Continuous Improvment

Lastly, we further evaluate the model’s accuracy by backtesting it every two weeks in order to get a sense of how good the model is at predicting compared to the real price values we gather every day.

What Our Clients Have to Say

"I strongly recommend AgFlow to other traders or brokers dealing locally and willing to expand beyond their region, as such market data and information are not easy to get for smaller players."

Nikola Sindjic

Trader, Agroglobe


Maximize Your ROI

We show you the daily transaction summaries received from our global network of 120+ brokers & traders. What you get is the actual market value of 6,000+ Grains, Oilseeds, and Vegoils quotes every day and 9,000+ freight rates weekly to ensure that you have the best execution!

Risk Management

Control Market Risk

Gain a consolidated view of risks and build high-level risk models thanks to AgFlow, get the full picture of your market and quantify the risks that your company’s activity can generate and reduce their impact.


Optimize Sourcing Costs

Every day, we collect raw data from our proprietary network of 130+ brokers & traders, structure it and aggregate and standardize it. Access thousands of quotes and import/export data on a daily basis in order to benchmark your prices and follow the demand.