/

CuspAI

  • What it is:CuspAI is a frontier AI company building a search engine for materials that accelerates discovery of breakthrough molecules for clean energy and sustainability using generative AI and simulations.
  • Best for:Large materials R&D organizations, Automotive OEMs (batteries, lightweight materials), Battery and energy storage developers
  • Pricing:Starting from Custom enterprise pricing
  • Rating:78/100Good
  • Expert's conclusion:CuspAI is most suited for climate focused organizations and large enterprises willing to make significant investments in AI-accelerated materials discovery for carbon capture, water treatment, etc. where time and cost savings are critical to their business needs.
Visit website
Reviewed byMaxim Manylov·Web3 Engineer & Serial Founder

What Is CuspAI and What Does It Do?

CuspAI is an AI pioneer, utilizing advanced AI techniques to improve materials science; it provides a materials search engine to rapidly identify and develop new materials. CuspAI uses a combination of generative AI, deep learning, and molecular simulation to decrease the time to discover new materials from thousands of years down to a few months. CuspAI has a diverse group of highly qualified world class researchers in AI, chemistry and engineering.

Active
📍Cambridge, United Kingdom
📅Founded 2024
🏢Private
TARGET SEGMENTS
Materials ScienceSustainable EnergyRenewable EnergyChemical DevelopmentAdvanced Manufacturing

What Are CuspAI's Key Business Metrics?

📊
$30M
Total Funding Raised
📊
Pre-Seed
Funding Stage
🏢
20-50
Employee Count
📊
Lightspeed Venture Partners
Lead Investor

How Credible and Trustworthy Is CuspAI?

78/100
Good

CuspAI has high levels of technical credibility via leading researchers and AI advisors; however, CuspAI is still a relatively immature startup founded in 2024. The early stage funding and pre-seed status indicates that CuspAI will have growth opportunities, however, CuspAI has little to no marketplace validation.

BREAKDOWN
Product Maturity65/100
Company Stability75/100
Security & Compliance70/100
Team Expertise95/100
Transparency80/100
Market Validation75/100
TRUST SIGNALS
Geoffrey Hinton ('Godfather of AI') serves as board advisorCo-founder Max Welling is world-renowned ML researcherChief Scientist is accomplished computational chemistFunding from Lightspeed Venture PartnersResearch teams across Cambridge and Amsterdam

What is the history of CuspAI and its key milestones?

2024

Company Founded

CuspAI was formed by Chad Edwards (CEO), Max Welling (CTO), Alyn Chad Edwards, and Geoffrey Hinton (board advisor).

2024

Pre-Seed Funding

CuspAI received $30M in pre-seed funding, from Lightspeed Venture Partners to create the AI platform for materials discovery.

2024

Research Team Assembly

CuspAI appointed a world-renowned computational chemist as its first Chief Scientist and has developed a research team across its Cambridge and Amsterdam locations.

What Are the Key Features of CuspAI?

✨
AI-Powered Materials Search Engine
The CuspAI system is a search engine for all possible molecules and materials that can potentially be synthesized, allowing users to generate and assess new materials on demand, rather than being limited to known materials.
✨
Generative AI for Material Design
The CuspAI system utilizes advanced generative AI to create novel materials with specific desired properties, which speeds up the discovery process from thousands of years down to months.
✨
Deep Learning and Molecular Simulation
The CuspAI system integrates deep learning with materials science and process design to predict and validate material properties accurately.
✨
On-Demand Material Generation
Users can input desired material properties into the CuspAI platform and receive candidate materials that meet their requirements.
🔗
Multi-Disciplinary Integration
CuspAI bridges the gap between AI technology and chemical/physical processes in materials science to combine these two disciplines.
✨
Materials Property Prediction
The CuspAI system predicts and assesses the properties of candidate materials before they are synthesized to minimize experimental cycle times and costs.

What Technology Stack and Infrastructure Does CuspAI Use?

Infrastructure

Research and development infrastructure with team members based across Cambridge and Amsterdam locations.

AI/ML Capabilities

Proprietary generative AI models with deep learning and molecular simulation capabilities designed specifically for materials science, leveraging the expertise of leading machine learning researchers.

Technical stack details limited as company is in early stage; information derived from official website and company descriptions of capabilities

What Are the Best Use Cases for CuspAI?

Materials Scientists and Chemists
Expedite new materials through the creation of novel materials and their evaluation for desired physical properties as needed, resulting in reduced R&D timeframes from years to months.
Battery and Energy Companies
Identify breakthrough materials for next generation battery technologies, renewable energy systems, and other sustainable energy opportunities to help meet global carbon reduction objectives.
Chemical Development and Manufacturing
Simplify the development and processing of chemicals and materials using AI-assisted discovery and process design similar to competitor Citrine Informatics' product line.
Aerospace and Advanced Materials Manufacturers
Design novel lightweight materials to provide specified performance attributes for aerospace, automotive, and high-technology manufacturers.
Sustainability-Focused Organizations
Develop breakthrough materials to enable sustainability and renewable energy projects to combat world-wide environmental issues.
NOT FORPharmaceutical Companies Requiring Small Molecule Discovery
Limited Application – The Platform has been designed specifically for the field of Materials Science as opposed to Pharmaceutical Drug Discovery, which requires a different set of validation and regulatory standards.
NOT FOROrganizations Requiring Immediate Production-Scale Results
Not Applicable – While the Platform will expedite discovery timelines for new materials, it does not eliminate the need for experimental verification of the synthesized materials prior to production.

How Much Does CuspAI Cost and What Plans Are Available?

Pricing information with service tiers, costs, and details
☐Service$CostℹDetails🔗Source
Enterprise Platform AccessCustom enterprise pricingB2B platform for AI-driven materials discovery. Contact sales for quotes based on research scope and compute needs.Company website and funding announcements
Enterprise Platform AccessCustom enterprise pricing
B2B platform for AI-driven materials discovery. Contact sales for quotes based on research scope and compute needs.
Company website and funding announcements

How Does CuspAI Compare to Competitors?

FeatureCuspAISchrödingerMaterials ProjectAspuru-Guzik Lab (XQTL)Recursion Pharmaceuticals
Core FunctionalityGenerative AI for materials (inverse design)Physics-based simulations + MLOpen database + simulationsAcademic ML models for crystalsAI for drug discovery (cellular imaging)
Pricing (starting price)Custom enterprise$10K+/year (typical)Free (academic)Research grantsPublic + enterprise partnerships
Free Tier AvailabilityNoLimited academicYesYes (publications)No
Enterprise FeaturesIndustry partnerships (Hyundai)SSO, audit logs (enterprise)——Yes (pharma partnerships)
API AvailabilityProprietary platformYes (via partnerships)Yes (open APIs)Research APIsEnterprise API
Integration CountExperimental pipelinesMultiple HPC/cloudOpen ecosystemAcademic toolsPharma workflows
Support OptionsDedicated partnerships24/7 enterpriseCommunityAcademicEnterprise support
Security Certifications—SOC 2, ISO——SOC 2, pharma compliant
Core Functionality
CuspAIGenerative AI for materials (inverse design)
SchrödingerPhysics-based simulations + ML
Materials ProjectOpen database + simulations
Aspuru-Guzik Lab (XQTL)Academic ML models for crystals
Recursion PharmaceuticalsAI for drug discovery (cellular imaging)
Pricing (starting price)
CuspAICustom enterprise
Schrödinger$10K+/year (typical)
Materials ProjectFree (academic)
Aspuru-Guzik Lab (XQTL)Research grants
Recursion PharmaceuticalsPublic + enterprise partnerships
Free Tier Availability
CuspAINo
SchrödingerLimited academic
Materials ProjectYes
Aspuru-Guzik Lab (XQTL)Yes (publications)
Recursion PharmaceuticalsNo
Enterprise Features
CuspAIIndustry partnerships (Hyundai)
SchrödingerSSO, audit logs (enterprise)
Materials Project—
Aspuru-Guzik Lab (XQTL)—
Recursion PharmaceuticalsYes (pharma partnerships)
API Availability
CuspAIProprietary platform
SchrödingerYes (via partnerships)
Materials ProjectYes (open APIs)
Aspuru-Guzik Lab (XQTL)Research APIs
Recursion PharmaceuticalsEnterprise API
Integration Count
CuspAIExperimental pipelines
SchrödingerMultiple HPC/cloud
Materials ProjectOpen ecosystem
Aspuru-Guzik Lab (XQTL)Academic tools
Recursion PharmaceuticalsPharma workflows
Support Options
CuspAIDedicated partnerships
Schrödinger24/7 enterprise
Materials ProjectCommunity
Aspuru-Guzik Lab (XQTL)Academic
Recursion PharmaceuticalsEnterprise support
Security Certifications
CuspAI—
SchrödingerSOC 2, ISO
Materials Project—
Aspuru-Guzik Lab (XQTL)—
Recursion PharmaceuticalsSOC 2, pharma compliant

How Does CuspAI Compare to Competitors?

vs Schrödinger

CuspAI utilizes Generative AI and Inverse Design to create potential material candidates at a significantly faster rate than Schrödinger’s Physics First Simulation Approach. CuspAI is focused on identifying breakthroughs in material discovery while Schrödinger is best suited for optimizing and facilitating regulatory compliance in workflows.

CuspAI for new materials moonshot projects; Schrödinger for validated industrial workflows.

vs Recursion Pharmaceuticals

CuspAI is focused on Materials Chemistry (MOFs, Batteries) while Recursion focuses on Biological Cellular Phenomics for Pharmaceutical Drug Discovery. Both are Enterprise Focused and have a strong foundation in AI; however they operate within distinct Scientific Domains.

CuspAI for hardware/materials innovation; Recursion for drug/biotech innovation.

vs Materials Project (Lawrence Berkeley)

Free Open Source Database vs. CuspAI’s Closed Loop AI Platform – Materials Project provides open access to its database; however, CuspAI’s Platform offers the ability to generate novel materials and is commercially scalable and capable of rapid deployment to production.

CuspAI for accelerating proprietary R&D; Materials Project for academic baseline research.

vs XQTL (Alan Aspuru-Guzik lab)

Academic Research using Machine Learning vs. CuspAI’s Commercially Scalable Enterprise Platform – XQTL has published SOTA Crystal Generators, however, CuspAI’s Platform includes integrated Experimental Validation Loops and Industry Partnerships to accelerate the deployment of new materials.

CuspAI for production-scale discovery; XQTL for cutting-edge academic research.

What are the strengths and limitations of CuspAI?

Pros

  • Top-tier founder team – most-cited researchers in both AI and chemistry with over 2M citations
  • Top-of-the-line, proprietary SOTA models – MOFGEN reaches a 49% VUN in performance as compared to Microsoft (10%) and Meta (16%).
  • Key enterprise partners – Hyundai Motor Group, Samsung Ventures, NVentures
  • Closed-loop discovery – combines simulation, experimentation and manufacturability in one process
  • Funding muscle – enables rapid growth with a $100M Series A at a $520M valuation
  • Advisors from industry leaders – Geoff Hinton, Yann LeCun, former BP CEO, current ASML CTO
  • Validated in real-world settings – active partnerships verify that the company is accelerating innovation

Cons

  • No enterprise pricing available – will not provide public pricing, trials or self-service options
  • Product still in early stages – funded Series A startup – has not yet proven itself at commercial scales
  • Limitations to domain-specificity – while the company is focused on chemistry and materials, it does not offer general AI capabilities
  • Capabilities are opaque – proprietary black-box models have been released without publication of any additional benchmark testing beyond MOFs
  • Expected high-cost enterprise pricing model – indicates that the company's deep technology is expensive
  • Long sales cycles – requires establishing a custom relationship with the customer versus providing an immediate SaaS experience
  • Risk associated with translating academic to commercial – converting research-based innovations into dependable products

Who Is CuspAI Best For?

Best For

  • Large materials R&D organizations — Inverse-design capabilities accelerate development of novel materials beyond standard screening methods
  • Automotive OEMs (batteries, lightweight materials) — Proven partner with Hyundai demonstrates that the company can accelerate the development of automotive materials
  • Battery and energy storage developers — Focuses on solid state electrolytes, power electronic devices, and modeling degradation
  • Electronics manufacturers (magnets, coatings) — Develops materials which do not use rare earth metals – thermal materials, corrosion-resistant materials
  • Water treatment and sustainability teams — Developing PFAS filtration systems, desalination membrane systems and sustainable polymer systems.

Not Suitable For

  • Small R&D teams or startups — The proprietary pricing model and sales process is a barrier to entry for companies working with limited budgets; CuspAI is an example of how using open source tools such as the Materials Project may provide solutions to these issues.
  • Academic researchers needing immediate access — CuspAI has chosen to use a proprietary platform over free academic alternatives to generate new molecular candidates and published models for their predictions and Materials Project.
  • Companies needing off-the-shelf SaaS — CuspAI has also been restricted by the need for custom enterprise deployments which may limit the number of customers they are able to onboard; companies who require faster deployment may be more likely to choose Schrödinger or Atomwise.
  • Biology/pharma drug discovery teams — While CuspAI is focused on designing molecules for the field of materials chemistry and not the field of biology, companies working on developing drugs through computational means will find Recursion or Insilico Medicine to be more suitable for their needs.

Are There Usage Limits or Geographic Restrictions for CuspAI?

Access Model
Enterprise customers and strategic partners only
Public Availability
No public platform, API, or self-serve access
Trial Access
Not available - contact sales for demos
Geographic Availability
UK-based (Cambridge) with EU operations (Amsterdam, Berlin)
Target Applications
Materials chemistry only - MOFs, batteries, coatings, membranes
Compliance Certifications
Free Tier

Is CuspAI Secure and Compliant?

Enterprise Security StandardsIndustry partnerships (Hyundai, Samsung) imply enterprise-grade security for sensitive IP.
Data ProtectionProprietary datasets and experimental pipeline integrations with access controls.
Research IP ProtectionDesigned for confidential materials R&D with Fortune 500 partners.
AI Model SecurityProprietary SOTA models protected for competitive advantage.

What Customer Support Options Does CuspAI Offer?

Channels
Dedicated account teams for enterprise clientsAccess to world-class advisors for strategic partners
Hours
Business hours across UK/EU timezones
Response Time
Enterprise partnership timelines
Specialized
CTO Max Welling and deep tech entrepreneur team provide strategic guidance
Business Tier
Dedicated success teams for strategic industry partnerships (Hyundai, Samsung)

What APIs and Integrations Does CuspAI Support?

API Type
Not publicly documented. CuspAI operates as a materials discovery platform with integration through partnerships rather than public API endpoints.
Integration Model
Enterprise partnerships and strategic collaborations (e.g., Meta FAIR team, Kemira). Integrations are customized for specific use cases like carbon capture material optimization and PFAS removal.
Use Cases
Materials design and optimization, carbon capture sorbent discovery, water treatment compound development, hydrogen storage material design. Can accelerate discovery from years to months.
Documentation
Limited public technical documentation. Company website and press materials provide overview of capabilities but detailed API documentation not publicly available.
Developer Access
Enterprise-focused. Requires direct partnership or enterprise engagement. No public developer portal or sandbox environment documented.

What Are Common Questions About CuspAI?

CuspAI uses Generative AI combined with Physics-Based Simulations to identify and rank molecular structure candidates. This allows CuspAI to avoid the "trial-and-error" approach commonly used when searching for new materials. Users simply input the properties they want their material to have and CuspAI identifies all possible candidates, and provides them to the user for testing and validation. This method allows CuspAI to reduce the time it takes to discover new materials from years down to months. Furthermore, this method has proven to work with approximately a 90% success rate.

CuspAI primarily focuses its efforts on climate and sustainability applications, which include but are not limited to: Carbon Capture, reduced cost per ton to below $150; Removal of Per- and Polyfluoroalkyl Substances from drinking water; Hydrogen Storage; and Gas Separation.

CuspAI can be applied to any materials discovery challenge where the discovery of a molecular structure is required to produce a material with specific molecular properties.

CuspAI currently works with several industry leading organizations, which include Meta's Fundamental AI Research team, where CuspAI is optimizing carbon capture for Meta, and Kemira, where CuspAI is partnering to remove PFAS and treat water in industrial processes. These partnerships demonstrate that CuspAI's technology has real world application and are validating the company's platform.

CuspAI's focus on practical deployment requirements for its technology along with its emphasis on sustainability applications, combined with its partnerships with large companies such as Meta and Kemira, differentiate CuspAI from competitors such as Orbital Materials and XtalPi.

CuspAI was founded by Chad Edwards (CEO) along with Max Welling, an AI pioneer and former Distinguished Scientist and Vice President of Microsoft Research and Qualcomm. The founders are among the most frequently cited researchers in AI, chemistry, and engineering.

CuspAI received $30 million in seed capital led by Hoxton Ventures (announced in mid-2024), and has since received an additional $100 million in Series A funding. The company should have sufficient funds to scale its operational activities and research & development efforts.

The AI-based materials science market is expected to grow from $16.55 billion in 2024 to $84.03 billion by 2033. In addition, the global carbon removal market is estimated at $1 trillion by 2030, which CuspAI could take advantage of with lower production costs.

Is CuspAI Worth It?

CuspAI represents a revolutionary method of discovering new materials that utilizes generative AI to shorten time lines for developing these materials while reducing costs associated with developing the necessary climate technologies. CuspAI has a number of favorable characteristics including a high quality technical base, partnerships with leading industry organizations, and a focus on implementing products as opposed to simply developing them, which will provide CuspAI with a competitive advantage in a growing market.

Recommended For

  • Chemical and materials companies focusing on sustainability (e.g., Kemira).
  • Climate technology companies working on carbon capture or removal.
  • Companies involved in water treatment and environmental remediation.
  • Technology companies focused on achieving their own climate goals (e.g., Meta).
  • Energy and industrial companies that seek to optimize the performance of their materials.
  • Venture investors that invest in climate-related and sustainable ventures.

!
Use With Caution

  • Companies that require immediate production-scale results — the company is currently refining the platform to ensure that it can meet the needs of large-scale manufacturers.
  • Organizations that do not have either partnerships or enterprise budgets — it appears that the platform will only be accessible to select organizations in a collaborative environment.
  • Companies that require established business-to-business relationships — CuspAI is a relative newcomer having launched in 2024.
  • Organizations that are hesitant to utilize materials designed using AI in regulated industries — CuspAI may need to obtain compliance validation before being able to supply these organizations with their products

Not Recommended For

  • Small businesses that do not have a budget to pursue sustainability initiatives -- enterprise pricing and partnerships
  • Organizations seeking pre-existing, off-the-shelf software solutions -- CuspAI operates through custom partnerships
  • Non-climate applications outside of CuspAI's current focus areas
  • Businesses that prefer working with established vendors that have decades of experience
Expert's Conclusion

CuspAI is most suited for climate focused organizations and large enterprises willing to make significant investments in AI-accelerated materials discovery for carbon capture, water treatment, etc. where time and cost savings are critical to their business needs.

Best For
Chemical and materials companies focusing on sustainability (e.g., Kemira).Climate technology companies working on carbon capture or removal.Companies involved in water treatment and environmental remediation.

What do expert reviews and research say about CuspAI?

Key Findings

CuspAI has achieved a 90 percent success rate in its materials discovery efforts, and has reduced the timeframe of these efforts from years to months; this positions CuspAI at the front of the pack of AI driven materials science. CuspAI has obtained significant validation from its partnerships with Meta and Kemira; has raised over $130 million dollars in funding; and is operating in a rapidly expanding marketplace expected to grow to $84 billion by 2033. CuspAI's highly skilled technical team of world class AI researchers provides a competitive advantage within a field that is experiencing intense competition from DeepMind backed Orbital Materials and existing players such as Dassault Systèmes.

Data Quality

Excellent — comprehensive public information from official CuspAI website, TechCrunch reporting, Kemira partnership announcement, and multiple investor/analyst articles. Funding details and partnership information well documented. Some internal metrics and pricing details not publicly available due to enterprise-focused business model.

Risk Factors

!
CuspAI is a young organization (founded in 2024) and has a very limited history of successful production deployments.
!
CuspAI is operating in a highly competitive marketplace with well funded competitors including Orbital Materials ($16 million series A round) and XtalPi (valuation of $2.5 billion).
!
CuspAI will need to obtain validation through actual commercial use -- all of CuspAI's partners are currently in the early stages of the partnership process.
!
Volatility associated with generative AI and the rapid evolution of models may negatively impact CuspAI's competitive position.
!
CuspAI's enterprise partnership model results in customer concentration risk.
!
CuspAI must demonstrate scalability of the manufacture of materials generated through the use of AI designs at an industrial level.
Last updated: February 2026

What Additional Information Is Available for CuspAI?

Founder Story

CuspAI has been founded by Chad Edwards and Max Welling who is a leading figure in the area of Artificial Intelligence. Max Welling has held the position of Distinguished Scientist & VP at both Microsoft Research and Qualcomm, and has also taught at the University of Amsterdam. This team has world-class experience in AI, Chemistry and Engineering which places the organization at the forefront of computational materials science.

Key Partnerships

Meta's Fundamental AI Research (FAIR) group are working together with CuspAI to optimize carbon capture materials and have received an endorsement of this partnership from Yann LeCun, Meta's VP & Chief AI Scientist. Kemira, one of the world's largest chemical companies has taken CuspAI and has put it into their R&D pipeline for the removal of PFAS in water and water treatment applications, giving access to the $12.5 billion dollar water treatment market.

Market Positioning

CuspAI is operating in an AI-driven materials science market that is growing at 19.8%, while the overall AI sustainability market will grow from $16.55B in 2024 to $84.03B in 2033. As a result of the Paris Agreement and the Inflation Reduction Act, CuspAI is benefiting from a $1 trillion global carbon removal market.

Funding & Valuation

Raised $30M Seed Round from Hoxton Ventures in mid-2024 and a subsequent $100M Series A round. CuspAI is well capitalized and is positioned to accelerate its growth and expand its research and development. It reflects investor confidence in the potential of the AI-driven materials discovery market.

Competitive Landscape

Competitors include Orbital Materials, founded by the founders of DeepMind with $16M Series A, XtalPi with $2.5B valuation, and Deep Principle. Simulation software companies such as established simulation software companies Dassault Systèmes and Schrödinger are also competitors to CuspAI. However, CuspAI is differentiated from other organizations in the space through its focus on sustainability and partnerships with industries.

Technical Approach

CuspAI combines Generative AI with Quantum Chemistry and Physics-Based Simulations to act as a Search Engine for Materials for users. Unlike traditional methods of materials design where users test physical prototypes, CuspAI allows users to design molecules to specifications and evaluate thousands of configurations in silico simultaneously.

Cost Impact

The partnership provides a path toward reduced costs: the goal of reducing costs associated with carbon capture to below $150 per ton is key to global adoption; and solutions for removing PFAS from water treatment provide a new market for the use of climate technology to improve speed and efficiency.

What Are the Best Alternatives to CuspAI?

  • •
    Orbital Materials: An AI-driven platform for discovering materials was founded by DeepMind alumni. It has the same generative AI approach to designing materials as CuspAI, but it has a much broader scope that includes climate. Raised $16 million in a series A funding round. This platform has received funding that is somewhat narrower than CuspAI but is backed by a lineage of deep learning. Therefore, this platform would be best for organizations that want to leverage the expertise of DeepMind alumni but do not require the significant strategic partnerships needed by CuspAI.
  • •
    XtalPi: XtalPi is valued at $2.5 billion and uses AI to discover and simulate materials. XtalPi uses computational chemistry and drug discovery in addition to materials discovery. Although XtalPi is more established and operates on a larger scale than CuspAI, it is also less focused on sustainability and climate-related applications. Therefore, XtalPi would be best for pharmaceutical and materials companies looking for broad computational capabilities.
  • •
    Dassault Systèmes: 3D Systems is an established provider of enterprise software that uses simulation and modeling tools for materials science. In addition to providing simulation tools, 3D Systems provides its BIOVIA suite of molecular simulation tools. Because of the larger scale and enterprise integration of 3D Systems, it is likely to have higher costs than CuspAI. However, because of its large scale and the fact that it has been used in many large organizations, 3D Systems may be a good choice for organizations with existing Dassault infrastructure.
  • •
    Schrödinger: XtalPi is a platform that uses computational chemistry and drug discovery to model compounds and predict their behavior. XtalPi is based on physics-based simulations and machine learning. XtalPi is a more mature platform than CuspAI and has a longer history of operation. However, XtalPi is not specifically designed to develop materials related to climate. Instead, XtalPi has a broader life sciences focus. Therefore, XtalPi would be best suited to pharmaceutical and biotech companies that are primarily interested in developing drugs.
  • •
    Deep Principle: AI-based discovery startup focused on electrocatalysts & battery materials, smaller-scale & less advanced in energy applications compared to CuspAI. Best fit for battery and energy storage companies. (deepprinciple.com)

Materials Discovery Performance Metrics

years to months discovery timeline reduction
Time-to-Discovery Reduction
90 % viable materials generated
Success Rate
49 % valid, unique, novel structures
MOFGEN VUN Rate
<150 $/ton CO₂
Carbon Capture Cost Reduction
vast numbers novel structures per query
Candidate Evaluation Speed

Computational Modeling & Simulation Features

Generative AI Material Design

Generate corresponding material composition by utilizing a reverse engineering approach to discover desired properties.

MOFGEN Autoregressive Transformer

Currently best model for Metal-Organic Frameworks (MOFs), at a 49% VUN rate, has exceeded results from Microsoft (10%), and Meta (16%).

Physics-Based Simulations

Deep learning, combined with Quantum Chemistry, is used to rank molecular structures based on their predicted properties.

Property Specification Engine

Users enter the specific target attributes such as; Thermal Stability, Electrical Conductivity, CO2 Selectivity, Mechanical Properties, etc.

Multi-Scale Modeling

Full end-to-end coverage from atomic/molecular design through to macro level process manufacturability.

Rapid Structure Evaluation

Like a search engine, it evaluates thousands of new structures in silico.

Data Integration & Standards Compliance

Proprietary Model Stack
Fully proprietary generative models covering complete materials discovery lifecycle
Property Database Access
Search engine accessing vast materials property space for candidate generation
Real-Time Candidate Generation
On-demand evaluation of novel structures matching specified performance criteria
Industrial Deployment Focus
Designed for regeneration efficiency and compatibility with existing systems

Industry-Specific Materials Discovery Use Cases

Industry SectorPrimary Use CaseMaterials FocusKey Performance TargetDiscovery Timeline
Climate TechCarbon capture and storageMolecular sponges, metal-organic frameworks (MOFs)CO₂ selectivity, <$150/ton capture costmonths
Clean EnergyBattery materials optimizationAdvanced cathode/anode materialsImproved performance and longevitymonths
Water PurificationPFAS forever chemicals removalSelective absorption materialsHigh efficiency pollutant removalmonths
Consumer ElectronicsAdvanced mechanical propertiesTailored structural materialsSpecific thermal/electrical conductivitymonths
Healthcare DevicesSpecialized material propertiesBiocompatible engineered materialsOptimized device performancemonths

Supported ML Frameworks & Technologies

Generative AIDeep LearningAutoregressive TransformersQuantum ChemistryMOFGEN (Proprietary)Physics-Informed ML

Compliance, Security & Reproducibility Certifications

Industrial Deployment ReadyDesigned for real-world manufacturing compatibility
90% Success Rate ValidationVerified viable material generation capability
Meta FAIR PartnershipCollaborating on CO₂-optimized MOFs
Kemira PFAS ProjectAI-driven water purification materials
SOTA Model Performance49% VUN rate outperforming industry leaders

Self-Driving Laboratory & Automation Capabilities

In Silico High-Throughput Screening

Thousands of novel material candidates can be rapidly evaluated in silico prior to physical synthesis.

Property-Driven Discovery

Automatically generates materials that match exact performance requirements.

Industrial Process Compatibility

Materials designed to be compatible with current manufacturing systems.

Real-World Validation Pipeline

90% Success rate from computational prediction through to experimental verification.

Open-Source Tools & Community Accessibility

Platform Type
Proprietary enterprise platform
Funding Raised
$100M+ for platform acceleration
Key Partnerships
Meta FAIR, Kemira, leading climate tech collaborators
Competitive Positioning
Outperforms Microsoft (10% VUN), Meta (16% VUN) models
Target Markets
Climate tech, clean energy, industrial manufacturing
Founders
Chad Edwards (Google/BASF), Max Welling (AI Professor)

Expert Reviews

📝

No reviews yet

Be the first to review CuspAI!

Write a Review

Similar Products

XtalPi
xtalpi.com
Pheronym
pheronym.com