MAGOS – collaboration of Neural Networks.

Tuned for predicting future events with high accuracy.


Our vision is to combine highly accurate forecasting ability of Neural Networks with blockchain technology. MAGOS fund, managed by AI and supervised by our team, will be deployed on the Ethereum platform. Initially operating on prediction markets and sportsbooks, it will evolve and expand into other fields, such as Digital Asset Management, where the forecasts of MAGOS will be applicable and valuable.


What is a Neural Network?

A Neural Network can be described as a software simulation of a biological brain, with a purpose to learn to recognize patterns in big data. Once the network has been trained on data samples, it can make predictions by detecting similar patterns in future data. Because of it’s ability to learn, Neural Network can be called true Artificial Intelligence, or AI.

What is MAGOS?

MAGOS is a complex forecasting model based on a system of five Neural Networks. Each network has its own purpose and their powers combined allows to predict outcomes of events with high accuracy, outperforming the crowd. The backbone of MAGOS is modular architecture that supports upgrades and development of the new features.

What are the results?

The open test of MAGOS started on February, 2017 and lasted 4 months. A third-party site with the open tracking of stats was used. The forecasting of e-sports – one of the largest growing betting markets – was chosen as a testing method.

The results have shown that MAGOS is able to hold a significant edge over the competition, gaining net positive returns after each month of testing, with the overall ROI (Return on Investment) of 28%.

For the more detailed breakdown of testing and results, please read our bluepaper.

Our vision

MAGOS will be generating profits by applying forecasting to various domains on multiple platforms. All the profits will be transferred into the Ethereum smart contract with the following distribution:

  • 85% to the token holders
  • 10% re-invested back into the MAGOS fund
  • 5% used for the operating expenses

The new token functionality in the MAGOS:Initiate release will allow token holders to vote on profit distribution changes.

At the beginning stages, the MAGOS will operate on sportsbooks and conventional prediction markets.

In early 2018, MAGOS will expand to Decentralized Ethereum-based prediction markets, like Gnosis, Augur, and Stox.

Going into the second half of 2018, and the release of a Melonport, a new decentralized version of MAGOS fund will be deployed on the Ethereum blockchain. By using Melon protocol, we will be able to setup a fully auditable and transparent fund, with the open performance tracking. This type of a fund would allow to translate the accuracy of MAGOS forecasts into the reputation growth of the fund in the prediction market and asset management space.

Our main focus in 2019 will be the release of Oraculum portal. It will allow everyone to purchase limited usage rights to MAGOS, and apply it’s forecasting power to the events of their choice. All payments will go into the profit smart contract and distributed accordingly.


What are MAG token and what's their purpose?

MAG is ERC-20 token on Ethereum blockchain. It can be used in both active and passive way. Token holders will receive dividends from all the fund operations, and can use the token voting mechanism to change various parameters within MAGOS, including profit distribution.

What is the maximum token supply?

The maximum supply of MAG tokens is 50,123,377. No more tokens will ever be created after the sale. For more information please refer to the details of the sale.

How can i purchase tokens?

Tokens are purchased by sending ETH to the token sale address. Please visit the token sale page if you wish to participate in MAG token sale.


Ante Magnusson

CEO, Co-Founder

Entrepreneur and a cryptocurrency enthusiast. Before becoming obsessed with deep learning and neural networks, Ante was a financial advisor and a trader who specialised in algorithmic trading, game theory, and technical analysis.

Andreas Theiss

CTO, Co-Founder

Full-stack data scientist with a strong background in mathematics and statistics. Andreas is an expert in predictive modeling and holds a MSPA (The Master of Science in Predictive Analytics) from Northwestern University. His current research interests are deeply rooted in using Artificial intelligence for forecasting a wide spectre of events.

Daniel Kim

Data Engineer

Student of data engineering with a hacker mindset and a passion for robotics. A wizard of data mining and creative data visualisation.

Isaac Welch

Solidity Developer

IT Security specialist and a Junior Smart Contracts Developer.

Christopher Davison

Sports Statistical Analyst

Seasoned sports statistician and consultant. Christopher has extensive experience in discovering new methods of using and presenting sports statistics.

Kim Seung Ho

E-Sports Advisor

Early bitcoin adopter and a professional handicapper with 12 years of experience. Kim Seung Ho is a master in a field of e-sports and a generator of endless ideas when it comes to finding and extracting new edges from e-sports data.

Alessandra Taylor

Marketing Advisor

Blockchain evangelist since 2015, business and marketing consultant with keen interest in decentralized infrastructure and prediction markets.

Mentioned on







Sign up for our newsletter