What Atombeat does

Atombeat is a privately held AI-driven computational drug discovery company founded in 2021 and headquartered in New Jersey, USA, with teams operating across the United States and China. The company serves pharmaceutical companies, biotechnology firms, and academic research institutions by providing cloud-based computational tools for hit discovery and lead optimization across small molecule, peptide, and novel conjugate programs. Its technology stack combines AI, physics-based modeling (notably Free Energy Perturbation), and high-performance computing into a cloud platform called Hermite®, with specialized modules including Uni-FEP, Uni-Dock, Uni-QSAR, Uni-Mol Docking, Uni-Fold, and RiDYMO®, the latter focused on 'undruggable' targets using Reinforced Dynamics.

The underlying platform integrates proprietary foundation models such as DPA-2 for QM-level energy surfaces and Uni-Mol for 3D molecular representation, with Uni-Dock capable of screening 100 million ligands in 8 hours via GPU acceleration on the NVIDIA BioNeMo platform. Uni-FEP achieves ~1.0 kcal/mol RMSE on a 376-target, 6,341-ligand benchmark, enabling chemically accurate binding affinity predictions, while the peptide design platform explores over one trillion candidate compounds built from 1000+ natural and non-natural amino acids.

Atombeat monetizes through three streams: subscription-based access to the Hermite® cloud platform (enterprise/custom pricing on multi-year contracts), fee-for-service or milestone-based drug discovery programs including end-to-end SM/Peptide/XDC hit discovery, and a computational FTE model that embeds dedicated AI, computational chemistry, and biology experts into customer workflows with flexible month-to-month billing. Go-to-market is enterprise field sales with demand generation via major industry conferences (LSX World Congress USA, Boulder Peptide Symposium, Discovery on Target) and technical workshops. Notable strategic relationships include a July 2025 partnership with CRDMO BioDuro to co-develop an AI-powered peptide drug discovery platform, and integration with NVIDIA BioNeMo for ultra-large virtual screening.

Name

Atombeat

Legal name

Atombeat Inc.

Website

https://atombeat.com

Company type

Private

Founded year

2021

Operating status

Operating

Headcount range

1–10 employees

Short description

Atombeat is a private, AI-native computational drug discovery company (founded 2021, New Jersey HQ) that sells cloud-based hit discovery and lead optimization software (Hermite®, Uni-FEP, Uni-Dock, RiDYMO®) alongside fee-for-service and computational FTE engagements to pharmaceutical companies, biotechs, and academic researchers globally.

Ownership category

akta.pro rank

Product category

Computational Drug Discovery Software

NAICS

Research and Development in Biotechnology (except Nanobiotechnology) (541714), Software Publishers (5132)

SIC

Services-Commercial Physical & Biological Research (8731)

akta.pro primary industry

AI/ML Platforms for Drug Discovery & Experiment Optimization (HLAGAJAO)

akta.pro secondary industries

Cheminformatics & Molecular Modeling / CADD (HLAGAJAE), Drug Discovery (Target ID/Validation & Hit Discovery) (HLAIALAA), Target Identification & Validation Platforms (omics-driven, chemoproteomics, genetic validation) (HLAAAIAI)

Keywords

• Computational drug discovery
• AI drug design
• Molecular docking software
• Free energy perturbation
• Peptide drug discovery

Headquarters

HQ city

Jersey City

HQ country

United States

HQ region

North America

Offices3 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

Personnel, Technology or R&D, Marketing or Sales, Operations, Infrastructure

Revenue model

1. Software Platform (SaaS): Cloud-based Hermite® platform offered as subscription or usage-based pricing for computational drug discovery tools including Uni-FEP, Uni-Dock, and Uni-QSAR.
2. Drug Discovery Services: Fee-for-service or milestone-based drug discovery services including SM/Peptide/XDCs hit discovery, expertise in 'undruggable targets', and computational FTE (full-time equivalent) dedicated experts for end-to-end computational workflow support with month-by-month billing.
3. Computational FTE: Dedicated computational experts in AI, computational chemistry, and biology providing end-to-end computational workflow support with flexible month-by-month billing.

Pricing tiers

Go-to-market motion2 records

Distribution channels3 records

Marketing channels6 records

Core offering

Atombeat develops cloud-based, AI- and physics-based computational drug discovery platforms (Hermite® and RiDYMO®) that enable hit discovery, lead optimization, and design of small molecules, peptides, and novel conjugates. The company also provides fee-for-service and computational FTE (full-time equivalent) drug discovery services, including specialized expertise for tackling 'undruggable' targets across pharmaceutical and biotechnology customers.

Product overview

Atombeat offers a platform-plus-modules architecture consisting of the Hermite cloud-based platform as the central interface, with specialized modules including Uni-FEP (free energy perturbation), Uni-Dock (GPU-accelerated virtual screening), Uni-QSAR (molecular property prediction), Uni-Mol Docking (3D binding pose prediction), and RiDYMO (dynamics-based hit discovery). The platform combines AI, physics-based modeling, and high-performance computing to democratize computational drug discovery, enabling researchers to explore over one trillion potential peptide compounds and accelerate hit discovery and lead optimization across small molecule, peptide, and novel conjugate programs.

Differentiator

Problem solved

Functional benefit

Brands

• Hermite®: Cloud-based, browser-accessible tools for hit discovery and lead optimization, powered by AI, physics and high-performance computing
• RiDYMO®
• Uni-FEP
• Uni-Dock
• Uni-Mol
• Uni-QSAR
• RiD (Reinforced Dynamics)

Products and services

• Hermite Platform Cloud-based, browser-accessible computational drug discovery platform for hit discovery and lead optimization, powered by AI, physics, and high-performance computing. Integrates modules including Uni-FEP, Uni-Dock, Uni-QSAR, and RiDYMO for pharmaceutical and biotechnology researchers.
• RiDYMO Platform Dynamics-based hit discovery and optimization platform that leverages reinforced dynamics to unlock challenging targets and accelerate early discovery of small molecule, macrocyclic peptide, and novel conjugate therapeutics. Designed for pharmaceutical and biotech drug discovery programs targeting difficult or undruggable proteins.
• Uni-FEP Free energy perturbation module enabling protein-ligand affinity computation at chemical accuracy (±1 kcal/mol). Features automated workflow for protein-ligand system preparation, enhanced sampling methods (Terminal Flip Monte Carlo, Water Swap Monte Carlo), and bespoke force field optimization using DPA-2. Used by pharmaceutical and biotech researchers for lead optimization.
• Uni-Dock GPU-accelerated molecular docking module capable of screening 100 million ligands in just 8 hours. Integrated on NVIDIA BioNeMo platform for ultra-large virtual screening of compound libraries by drug discovery teams.
• Uni-QSAR Auto-ML tool for molecular property prediction including ADME/T properties, cell membrane permeability, and drug efflux prediction. Used for virtual screening and structure-activity relationship (SAR) exploration in drug discovery programs.
• Uni-Mol Docking 3D molecular representation learning framework for binding pose prediction, benchmarked against AlphaFold 3 as state-of-the-art model. Uni-Mol Docking v2 provides realistic and accurate binding pose prediction for drug discovery researchers.
• Uni-Fold Open-source platform for developing protein folding models, including Uni-Fold Symmetry for folding large protein complexes with symmetry constraints. Used by computational biology and structural biology researchers.
• Uni-GBSA Open-source web-based automatic workflow for performing MM/GBSA (Molecular Mechanics/Generalized Born Surface Area) calculations for binding affinity prediction in computational drug discovery workflows.
• DPA-2 Foundation Model In-house machine learning potential foundation model for reproducing QM-level energy surfaces of biological molecules. Used for bespoke torsion parameter optimization in Uni-FEP force field fitting workflow.
• Uni-RNA Universal pretrained models for RNA research, supporting RNA structure prediction and analysis for computational biology and drug discovery researchers working on RNA-targeted therapeutics.
• Drug Discovery Services (SM/Peptide/XDCs Hit Discovery) Fee-for-service or milestone-based drug discovery services including small molecule, peptide, and novel conjugate (XDC) hit discovery with specialized expertise in 'undruggable' targets. Delivered to pharmaceutical and biotech customers as project-based engagements.
• Computational FTE Services Dedicated computational experts in AI, computational chemistry, and biology providing end-to-end computational workflow support for pharmaceutical and biotech customers. Flexible month-by-month billing model with multi-year contract options.

Quantifiable outcome

• Screening 100 million ligands in just 8 hours using Uni-Dock
• +4 more outcomes

Named customers3 records

Segments4 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability8 records

Feature9 records

Partnerships

One partnership is on record.

• BioDuroflagshipStrategic or Co-development Partner · 30 July 2025Strategic collaboration to develop an AI-powered platform for accelerated peptide drug discovery. The partnership combines Atombeat's AI-driven design technology (capable of exploring over one trillion potential peptide compounds) with BioDuro's high-throughput synthesis and testing capabilities. The platform enables seamless end-to-end workflow from molecular design to preclinical candidate, delivering high-quality cyclic peptides with faster timelines and better economy across therapeutic areas. BioDuro is an Advent International portfolio company and CRDMO with 29-year track record, 7 sites across US and China, and 2,000+ employees.

Scale indicators9 records

Recent moves6 records

Expansion highlights5 records

Direct peers

• Insilico Medicine: Insilico Medicine is an AI-driven drug discovery company combining generative AI, biology, and chemistry platforms to discover novel therapeutics — directly comparable to Atombeat's AI/physics hybrid approach with a stronger pipeline orientation.
• Schrödinger: Schrödinger is the publicly traded leader in computational chemistry and physics-based drug discovery, offering FEP+, LiveDesign, and a broad platform overlapping directly with Atombeat's Uni-FEP, Uni-Dock, and Hermite offerings to pharma and biotech customers.
• OpenEye (Cadence Molecular Sciences): OpenEye, now part of Cadence, offers CADD software (OpenFEP, Orion, Floe) and cheminformatics tools targeting similar pharma and biotech computational chemistry users as Atombeat, with overlapping free-energy perturbation and virtual screening capabilities.
• Exscientia: Exscientia pioneered AI-driven drug design with its Centaur Chemist platform, partnering with pharma companies on AI-generated drug candidates. Competes with Atombeat for similar pharma AI-discovery service contracts and platform mindshare.
• Cresset: Cresset provides computational chemistry software and services including Flare for ligand-based and structure-based drug design, directly competing with Atombeat in the CADD/SBDD space serving medicinal chemists at pharma and biotech companies.

Broad incumbents

• Certara: Certara is a publicly traded biosimulation and drug development technology company offering software (Phoenix, Simcyp) plus services. Its broad portfolio across PBPK, PK/PD, and regulatory science overlaps with Atombeat's computational modeling customer base, though Certara focuses on later-stage development.
• Dotmatics: Dotmatics (backed by Insight Partners) offers scientific informatics platforms including GraphPad, SnapGene, and Dotmatics Studio for drug discovery R&D. While broader in scope, it competes with Atombeat for pharma R&D software budget and enterprise informatics deployments.
• Recursion: Recursion combines high-throughput wet-lab biology with AI/ML for drug discovery. While its wet-lab-first approach differs from Atombeat's pure-computational model, both target pharma AI-discovery partnerships and compete for similar biotech customer mindshare.

Emerging players

• Chemical.AI: Chemical.AI offers AI-powered retrosynthesis planning and organic chemistry automation tools. It addresses an adjacent workflow in computational chemistry, overlapping with Atombeat's cheminformatics customer base in pharma and academic drug discovery.
• Iambic Therapeutics: Iambic Therapeutics is an AI-driven drug discovery startup combining physics-based models with neural network potentials for small molecule and ADC design — a closer technological match to Atombeat's AI/physics hybrid, though smaller and more pipeline-oriented than Atombeat's platform-plus-services model.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights7 records

Customer concentration

• LinkedInlinkedin.com/company/atombeat-incorporated

Revenue estimate

Valuation estimate

Number of profiles

Profiles11 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Atombeat