
Numeraire is a cryptocurrency developed by a hedge fund firm. The company supplies cleaned, normalized, and obfuscated datasets to enable researchers to forecast stock market movements. By integrating data science, artificial intelligence, machine learning, and cryptography, the organization crowdsources effective machine learning models while maintaining strict data confidentiality.
Numeraire enables stock market trend prediction and trading decision-making by crowdsourcing data intelligence and analytics. This is accomplished through two user models: Numerai Tournament and Signals. Participants design machine learning models using provided training datasets and must stake NMR. Rewards are distributed according to the performance of each model.
The firm behind Numeraire was founded by Richard Craib in 2015. Craib holds degrees in mathematics and economics from Cornell University and previously worked as a data scientist at an asset management company. The platform’s white paper was released on February 20, 2017, followed by the Erasure Quant white paper in August 2019.
Since its launch in 2017, the project has been funded solely by NMR token sales. The original total supply was set at 21 million NMR, but in July 2019, the maximum supply was reduced to 11 million and the controlling key was destroyed. As a result, the NMR token contract can no longer be managed or upgraded. NMR is now a deflationary asset that cannot be minted, only burned.
The platform supports staking via two user models: Signal and Tournament. Participants who develop models offering reliable trading strategies and predictions are scored on performance and rewarded with NMR tokens.
Major competitors to Numeraire include QuantConnect, Quantiacs, and WorldQuant. These platforms also crowdsource machine learning models from data scientists to predict stock market trends. However, Numeraire stands out by leveraging smart contracts, creating a trustless system.
Numeraire combines artificial intelligence with the resilience and immutability of blockchain to predict stock market movements. Weekly tournaments using the Erasure protocol attract data scientists to build machine learning models and stake NMR.
Machine learning algorithm development is challenging for non-experts, which may restrict tournament participation to individuals with advanced computer science skills.
Machine learning based on Numeraire’s encrypted data could revolutionize various financial sectors. The emerging DeFi ecosystem, in particular, stands to benefit significantly from blockchain-integrated AI technologies.
The griefing mechanism in the Erasure protocol presents a challenge: buyers may not always be willing to pay to register dissatisfaction, potentially leading to a surge of spam predictions in the marketplace.
Numeraire fuses artificial intelligence and blockchain technology, delivering a novel approach to stock market prediction and analysis. Through crowdsourcing and sophisticated incentive structures, it aggregates advanced machine learning models to optimize hedge fund management. Despite notable advantages, challenges remain, including high entry barriers and increased competition. In the future, Numeraire’s expansion into decentralized finance and broader financial applications will be pivotal to its development.
NMR is the native token of the Numeraire platform, which utilizes AI and machine learning to crowdsource stock investment data. Operating on Ethereum, NMR has a capped supply of 11 million tokens. Users earn NMR by making accurate predictions.
NMR’s value lies in staking for hedge fund rewards. Investment decisions should reflect market conditions and individual risk tolerance.
On August 29, 2025, NMR dropped 1,163.31% in 24 hours, then surged 8,310.19% over seven days. Within a month, it climbed 8,352.67%, highlighting substantial price volatility.
The price increase is fueled by a $1 million buyback program, reduced supply, price momentum, and platform growth. With its relatively small market capitalization, NMR responds sharply to liquidity changes.











