Nova
Nova
[ ABOUT NOVA ]
NOVA is a decentralized engine for therapeutic discovery. Our approach transforms the quest for new therapeutics into a competitive, transparent, and efficient process. Running 24/7, NOVA functions as a back-to-back hackathon. Participants are encouraged to continuously develop and refine AI-driven approaches that make drug hunting in ultra-large chemical and biological universes more efficient. The result is a continuously improving platform designed to accelerate scientific innovation from computational discovery through downstream R&D.
NOVA is a decentralized engine for therapeutic discovery. Our approach transforms the quest for new therapeutics into a competitive, transparent, and efficient process. Running 24/7, NOVA functions as a back-to-back hackathon. Participants are encouraged to continuously develop and refine AI-driven approaches that make drug hunting in ultra-large chemical and biological universes more efficient. The result is a continuously improving platform designed to accelerate scientific innovation from computational discovery through downstream R&D.
PHARMA
CRISIS
DRUG DISCOVERY COST
$1 billion USD
10% Chance
SUCCESS RATE TO DEVELOP ONE DRUG
Time to Market
10+ YEARS
90%
OF DRUG SCAFFOLD REUSED
PHARMA
CRISIS
DRUG DISCOVERY COST
$1 billion USD
SUCCESS RATE TO DEVELOP ONE DRUG
10% Chance
Time to Market
10+ YEARS
OF DRUG SCAFFOLD REUSED
90%
Advanced virtual
screening can help, but
will require merging:
Insight
Humans
![](data:image/svg+xml,<svg display="block" role="presentation" viewBox="0 0 18 18" xmlns="http://www.w3.org/2000/svg"><path d="M 0 0 L 17.053 0 L 17.053 16.84 L 0 16.84 Z" fill="transparent" height="16.83991683991684px" id="zAfsI9PZJ" stroke-linecap="butt" stroke-linejoin="miter" stroke-miterlimit="10" stroke-width="0.69" stroke="rgb(255, 255, 255)" transform="translate(0.474 0.579)" width="17.05263157894737px"/><path d="M 0 16.571 L 16.781 0" fill="transparent" height="16.571175623700604px" id="S6pESBZB8" stroke-linecap="butt" stroke-linejoin="miter" stroke-miterlimit="10" stroke-width="0.69" stroke="rgb(255, 255, 255)" transform="translate(0.61 0.713)" width="16.78052368421053px"/><path d="M 16.78 16.571 L 0 0" fill="transparent" height="16.571175623700604px" id="GyqLGZBkm" stroke-linecap="butt" stroke-linejoin="miter" stroke-miterlimit="10" stroke-width="0.69" stroke="rgb(255, 255, 255)" transform="translate(0.61 0.713)" width="16.780495263157896px"/></svg>)
Scientific intuition
Domain expertise
Mechanistic insight
Problem definition
Research
Agents
Literature synthesis
Hypothesis generation
Empirical analysis
Experimental planning
Prediction
Machine learning
Screenings
Chemical exploration
Multi parametric optimization
Our approach
1
Open by design
NOVA's code, algorithms, and libraries are open source. Every breakthrough is shared, building on collective intelligence rather than siloed R&D.
2
Incentives aligned with progress
Participants are only rewarded for winning submissions, tying economic incentives directly to progress and efficiency.
4
Continuously improving
Winning molecules and algorithms raise the baseline, while wet lab results ground-truth the predictions.
3
Global Intelligence
Anyone can participate. NOVA coordinates machine learning, AI agents, and human expertise from around the world, rewarding performance rather than credentials or institutional affiliation.
5
Multiple therapeutic modalities
By screening small molecules and biologics together, NOVA tackles multiple modalities that can address a broader range of conditions.
6
Innovation in techniques not just outputs
Beyond surfacing high-scoring molecules and nanobodies, NOVA incentivizes novel search approaches.
7
Interdisciplinary innovation
Our competitions do not require domain expertise for participation and have already demonstrated how interdisciplinary approaches can beat traditional pharma techniques.
8
Scalable Scientific Infrastructure
NOVA is designed as a platform rather than a single drug program—an infrastructure layer capable of supporting many therapeutic targets, partners, and discovery campaigns simultaneously.
1
Open by design
NOVA's code, algorithms, and libraries are open source. Every breakthrough is shared, building on collective intelligence rather than siloed R&D.
2
Incentives aligned with progress
Participants are only rewarded for winning submissions, tying economic incentives directly to progress and efficiency.
3
Global Intelligence
Anyone can participate. NOVA coordinates machine learning, AI agents, and human expertise from around the world, rewarding performance rather than credentials or institutional affiliation.
4
Continuously improving
Winning molecules and algorithms raise the baseline, while wet lab results ground-truth the predictions.
5
Multiple therapeutic modalities
By screening small molecules and biologics together, NOVA tackles multiple modalities that can address a broader range of conditions.
6
Innovation in techniques not just outputs
Beyond surfacing high-scoring molecules and nanobodies, NOVA incentivizes novel search approaches.
7
Interdisciplinary innovation
Our competitions do not require domain expertise for participation and have already demonstrated how interdisciplinary approaches can beat traditional pharma techniques.
8
Scalable Scientific Infrastructure
NOVA is designed as a platform rather than a single drug program—an infrastructure layer capable of supporting many therapeutic targets, partners, and discovery campaigns simultaneously.
NOVA
3 competitions, running 24/7
1 system
NOVA
3 competitions, running 24/7
1 system
MODELS
SMALL MOLECULES
Nanobodies
Algorithms
MODELS
Drug development
How does it
work?
How does it
work?
STEP
1
define
challenge parameters including: chemical space, drug targets, and oracle model
By
METANOVA + partners
STEP
2
compete
identify and submit most promising molecules, nanobodies and/or search algorithms
By
Miners
STEP
3
score
score submissions using SOTA prediction model
By
Validators
STEP
6
retrain
integrate ground truth data to compound learning cycle
By
METANOVA
STEP
5
test
run wet lab validation
By
CRO + PARTNERS
STEP
4
shortlist
select top submissions from each challenge
By
METANOVA
STEP
develop + deliver
advance top assets through licensing, regulatory approval, and market entry through strategic partnerships + internally
Virtual drug
Screening at
Lightning speed
Virtual drug
Screening at
Lightning speed
Virtual drug
Screening at
Lightning speed
Value Stack
Discovery infrastructure that grows over time
Nova's competitive infrastructure generates a growing stack of assets for drug discovery
Value Stack
Value Stack
FAQ
NOVA is a decentralized drug discovery subnet on the Bittensor network. Traditional pipelines often take over a decade and cost billions per drug—our approach leverages global compute, talent, SOTA models, and a dynamic incentive system to drive efficiency, quality, and innovation.
What is NOVA and why does it matter?
NOVA is a decentralized drug discovery subnet on the Bittensor network. Traditional pipelines often take over a decade and cost billions per drug—our approach leverages global compute, talent, SOTA models, and a dynamic incentive system to drive efficiency, quality, and innovation.
What is NOVA and why does it matter?
There are 3 continuous machine learning competitions focused on small molecules, nanobodies, and chemical search algorithms. Each submission is evaluated by scoring criteria and oracle models specific to each task. The winning submissions for each competition is rewarded in $NOVA. This winner-takes-all approach rewards genuine innovation and continually improves the quality of submissions.
What does the “competition model” look like and how are winners determined?
There are 3 continuous machine learning competitions focused on small molecules, nanobodies, and chemical search algorithms. Each submission is evaluated by scoring criteria and oracle models specific to each task. The winning submissions for each competition is rewarded in $NOVA. This winner-takes-all approach rewards genuine innovation and continually improves the quality of submissions.
What does the “competition model” look like and how are winners determined?
Most AI platforms for drug discovery are centralized, limited by their own compute capacity and proprietary data. NOVA, on the other hand, distributes the workload across a global network of participants—miners, validators, and data scientists—who compete in real time. This collective intelligence, combined with on-chain transparency, transforms drug discovery into an open, merit-based process rather than a black-box service.
How does NOVA differ from traditional AI-driven drug discovery?
Most AI platforms for drug discovery are centralized, limited by their own compute capacity and proprietary data. NOVA, on the other hand, distributes the workload across a global network of participants—miners, validators, and data scientists—who compete in real time. This collective intelligence, combined with on-chain transparency, transforms drug discovery into an open, merit-based process rather than a black-box service.
How does NOVA differ from traditional AI-driven drug discovery?
Large-Scale Production Screenings Our Small Molecule and Nanobody competitions are analogous to large-scale industrial screening campaigns. Participants identify starting points that can advance through downstream computational analysis, experimental validation, and broader R&D. Discovery Algorithm R&D Lab Our Chemical Search Algorithm competition functions as an algorithm research laboratory. Rather than producing therapeutic candidates directly, participants develop new search strategies that make future discovery campaigns faster, more efficient, and better at exploring novel regions of chemical space. These advances improve the entire platform and benefit subsequent screening campaigns.
What is the traditional analogue for these competitions?
Large-Scale Production Screenings Our Small Molecule and Nanobody competitions are analogous to large-scale industrial screening campaigns. Participants identify starting points that can advance through downstream computational analysis, experimental validation, and broader R&D. Discovery Algorithm R&D Lab Our Chemical Search Algorithm competition functions as an algorithm research laboratory. Rather than producing therapeutic candidates directly, participants develop new search strategies that make future discovery campaigns faster, more efficient, and better at exploring novel regions of chemical space. These advances improve the entire platform and benefit subsequent screening campaigns.
What is the traditional analogue for these competitions?