Modella AI
Modella AI is a Boston-based biomedical AI company that develops multimodal foundation models and generative AI copilots for computational pathology, serving pharmaceutical oncology R&D and digital pathology platforms. Founded in 2024 as a Harvard/Mass General Brigham spinout, it was acquired by AstraZeneca in January 2026.
- Company typePrivate
- Founded2024
- HeadquartersBoston, United States
- Headcount1–10
- GTM typeB2B
- OfferingSoftware
What Modella AI does
Modella AI is a Boston-based biomedical AI company founded in 2024 as a spinout from the Mahmood Lab at Harvard Medical School and Mass General Brigham. The company develops multimodal foundation models and generative AI tools for biomedical imaging, with an initial focus on computational pathology. Its core product portfolio includes PathChat (a multimodal generative AI copilot for pathologists that reasons over pathology images and text for report summarization, differential diagnosis, and contextual question answering), PathChat DX (an advanced research-use version granted FDA Breakthrough Device Designation), Judith (an AI agent that automates complex biomedical image analysis workflows including cell and tissue segmentation and biomarker discovery), and proprietary foundation models (UNI, HIPT) pretrained on extensive histology image and image-text clinical datasets.
Underlying technology combines pathology foundation models with custom-trained multimodal large language models (MLLMs) to analyze high-resolution whole slide images alongside clinical text. Platform capabilities are validated through publications in Nature, Nature Medicine, Nature Biomedical Engineering, Nature Biotechnology, Nature Cancer, and Cancer Cell, and through FDA Breakthrough Device Designation for PathChat DX. The company pursued enterprise-level B2B licensing and multi-year research collaborations with pharmaceutical companies, with AstraZeneca as its flagship customer — beginning as a multi-year strategic partnership in July 2025 and culminating in a full acquisition by AstraZeneca announced at the J.P. Morgan Healthcare Conference in January 2026. Pre-acquisition revenue was generated through research-use platform licensing, strategic co-development partnerships, and academic/government grants, with all financial terms undisclosed.
In January 2026, Modella AI was acquired by AstraZeneca — described as the first acquisition of an AI firm by a major pharmaceutical company — and now operates as a wholly-owned subsidiary integrated into AstraZeneca's global oncology R&D organization. Pre-acquisition, the company had 1–10 employees and pursued a partnership-led, enterprise-focused go-to-market anchored in scientific credibility, peer-reviewed publications, and executive relationships in precision medicine, rather than self-serve product distribution.
Modella AI firmographics
Firmographics- Name
- Modella AI
- Legal name
- Modella AI, Inc.
- Website
- https://modella.ai
- Company type
- Private
- Founded year
- 2024
- Operating status
- Acquired
- Headcount range
- 1–10 employees
- Short description
- Modella AI is a Boston-based biomedical AI company that develops multimodal foundation models and generative AI copilots for computational pathology, serving pharmaceutical oncology R&D and digital pathology platforms. Founded in 2024 as a Harvard/Mass General Brigham spinout, it was acquired by AstraZeneca in January 2026.
- Ownership category
- akta.pro rank
Modella AI industry classification
Industry- Product category
- Computational Pathology AI
- NAICS
- In-Vitro Diagnostic Substance Manufacturing (325413), Scientific Research and Development Services (5417), Research and Development in Biotechnology (except Nanobiotechnology) (541714)
- SIC
- Surgical & Medical Instruments & Apparatus (3841), In Vitro & In Vivo Diagnostic Substances (2835)
- akta.pro primary industry
- Medical Imaging AI (HDAAAEAH)
- akta.pro secondary industries
- AI/ML-Enabled Biologic Discovery Platforms (computational design, in silico screening) (HLAAAIAE), SaMD – Imaging & Signal Analysis (AI/ML) (HLACANAG)
Keywords
Where Modella AI is headquartered
LocationHeadquarters
- HQ city
- Boston
- HQ country
- United States
- HQ region
- North America
Offices1 record
Markets served
Modella AI business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Personnel, Technology or R&D, Operations, Infrastructure, Marketing or Sales
Revenue model
- AI Platform Licensing & Research Collaboration: Modella AI generated revenue through multi-year research agreements and platform licensing deals with pharmaceutical companies for access to its multimodal AI foundation models. The AstraZeneca agreement (July 2025) exemplifies this model, providing access to foundation models for oncology R&D acceleration. Post-acquisition, revenue flows internally within AstraZeneca.
- Technology Licensing (Pre-Acquisition): Licensing of proprietary AI models (PathChat, Judith, foundation models) to healthcare institutions, digital pathology platforms, and life sciences companies. Financial terms of transactions were generally not disclosed.
- Strategic Partnership Revenue: Revenue from strategic co-development partnerships with pharma and biotech companies targeting clinical development, biomarker discovery, and patient outcome improvements. Modella AI collaborated closely with partners to integrate models into research pipelines.
Go-to-market motion1 record
Distribution channels3 records
Marketing channels6 records
Modella AI product offering
Product offeringCore offering
Modella AI builds multimodal foundation models and generative AI copilots for computational pathology. Its products — PathChat, PathChat DX (FDA Breakthrough Device Designation), and Judith — analyze high-resolution pathology images alongside clinical text to support diagnosis, biomarker discovery, patient stratification, and drug development workflows for pharmaceutical, biotech, and pathology research customers.
Product overview
Modella AI is a biomedical AI company that developed a portfolio of multimodal generative AI products centered on computational pathology. The core offerings include PathChat, a multimodal generative AI copilot for pathologists; PathChat DX, an advanced version with FDA Breakthrough Device Designation; and Judith, an AI agent for automating biomedical image analysis workflows. These products are powered by proprietary multimodal foundation models that integrate pathology images with clinical data. In January 2026, Modella AI was acquired by AstraZeneca, and its platform is now being integrated into AstraZeneca's global oncology R&D operations to accelerate clinical development and biomarker discovery.
Differentiator
Problem solved
Functional benefit
Brands
- PathChat: AI copilot for consulting on pathology cases. Summarize and create pathology reports from microscope, slide viewer, or smartphone. Includes PathChat 2, a multimodal generative AI model that reasons over pathology images and text for research and education use.
- Judith
Products and services
- PathChat Multimodal generative AI copilot for pathologists that enables intelligent interactions with digital slides, offering report summarization, image interpretation, differential diagnosis, morphological description, and contextual question answering. Accepts multiple high-resolution images interleaved with text for comprehensive assessment of consult cases through an interactive slide viewer.
- PathChat DX Advanced research-use-only version of PathChat granted FDA Breakthrough Device Designation. Integrates generative AI and multimodal analysis designed to assist pathologists in diagnosing complex cases with greater accuracy and efficiency. Demonstrated extremely high concordance with expert pathologist consensus reads when analyzing virtual H&E images from unstained tissue in a joint study with illumiSonics.
- Judith First research-use AI agent for biomedical image analysis that accelerates scientific discovery by customizing and automating complex workflows. Users describe tasks in natural language and Judith handles all complexities of building and interpreting AI models for biomedical imaging datasets, including cell and tissue segmentation in gigapixel pathology images, biomarker discovery, and whole slide image retrieval using foundation models.
- Multimodal Foundation Models Multimodal foundation models pretrained on extensive histology image and image-text datasets that provide powerful features from different data types for downstream tasks including disease detection, diagnosis, prognosis, and therapy prediction. Integrate visual information from pathology slides with textual clinical data for comprehensive analysis. Includes the proprietary UNI, HIPT, PORPOISE, CLAM, and TOAD model families developed at the Mahmood Lab.
Quantifiable outcome
- PathChat DX demonstrated extremely high concordance with expert pathologist consensus reads in a joint study with illumiSonics analyzing virtual H&E images from unstained tissue
- +2 more outcomes
Companies that use Modella AI
Customer profileNamed customers1 record
Segments3 records
Ideal customer profiles3 records
Modella AI technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- No
- API docs
- API detail
Core technology
AI maturity
App detail
Integration2 records
AI capability8 records
Feature5 records
Modella AI partnerships and signals
Strategic signalPartnerships
Two partnerships are on record, tiered flagship (core) and core.
- AstraZenecaflagship (core)Modella AI entered a multi-year strategic partnership with AstraZeneca in July 2025 to provide access to its state-of-the-art multi-modal AI foundation models for AstraZeneca's global oncology portfolio. The partnership was expanded into a full acquisition announced at the J.P. Morgan Healthcare Conference in January 2026, integrating Modella AI's generative and agentic AI platform into AstraZeneca's oncology R&D organization to accelerate clinical development, enhance biomarker discovery, and enable data-driven decision-making. Financial terms were not disclosed.
- TechcytecoreModella AI partnered with Techcyte as part of the Techcyte Fusion Partner Program, integrating its research-use only PathChat AI copilot into Techcyte's cloud-based digital pathology platform. The collaboration supports workflow efficiency and expands access to advanced AI-powered pathology tools for research, combining Modella AI's generative AI co-pilot capabilities with Techcyte's integrated digital pathology platform.
Scale indicators4 records
Recent moves5 records
Expansion highlights4 records
Modella AI competitors and assessment
Company assessmentDirect peers
- Lunit: Lunit develops AI for medical imaging, including pathology and radiology, with a focus on cancer detection and biomarker prediction. It competes with Modella in AI-driven pathology image analysis and pharma biomarker applications.
- Paige AI: Paige AI builds AI-based computational pathology software for cancer diagnosis and biomarker discovery, with FDA-authorized products and pharma partnerships. It is the closest direct competitor to Modella AI in foundation-model-driven pathology and shares overlapping academic lineage from the Mahmood Lab; Modella's CEO Jill Stefanelli was previously President and CBO at Paige.
- Proscia: Proscia provides a digital pathology platform with AI-powered applications for diagnosis and research, including integrations with AI model providers. It overlaps with Modella as both enable pathology AI workflows and Proscia's platform competes with embedded offerings like Modella's Techcyte integration.
- PathAI: PathAI develops AI-powered pathology tools for diagnosis and biopharma R&D, including biomarker discovery and clinical trial applications. It directly competes with Modella in applying foundation-model pathology to pharma oncology workflows.
- Ibex Medical Analytics: Ibex builds AI-driven pathology diagnostics for cancer detection, with clinically deployed products in hospitals and labs. It competes with Modella's diagnostic interpretation use cases in pathology, particularly as both pursue clinical-grade generative AI.
- Techcyte: Techcyte operates a cloud-based digital pathology platform and runs a partner program that integrates third-party AI like Modella's PathChat. It is both a partner and a peer, since its platform competes with other AI-enabled digital pathology offerings that bundle similar workflow capabilities.
- Owkin: Owkin applies AI to clinical trials, patient stratification, and biomarker discovery for pharma, with a federated learning platform across hospital networks. It overlaps with Modella's oncology R&D and clinical-trial use cases at the intersection of AI and pharma drug development.
Broad incumbents
- Tempus Labs: Tempus is a larger, publicly visible AI-powered precision medicine company combining molecular testing, data, and AI for oncology R&D and clinical decision support. It overlaps with Modella's biomarker discovery and oncology R&D use cases as a broad incumbent operating across the same customer base.
Market position
Strengths5 records
Weaknesses4 records
Competitive moat4 records
Key risks6 records
Key highlights7 records
Customer concentration
Modella AI social profiles
Digital presenceModella AI compliance and trust
Trust signalCompliance1 record
Modella AI financial estimates
Financial estimateRevenue estimate
Valuation estimate
Modella AI leadership team
Management profileNumber of profiles
Profiles8 records
Modella AI funding detail
Funding detailFunding overview
Funding rounds
Investors
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Modella AI M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Modella AI
What does Modella AI do?
Modella AI builds multimodal foundation models and generative AI copilots for computational pathology. Its products — PathChat, PathChat DX (FDA Breakthrough Device Designation), and Judith — analyze high-resolution pathology images alongside clinical text to support diagnosis, biomarker discovery, patient stratification, and drug development workflows for pharmaceutical, biotech, and pathology research customers.
Is Modella AI a public or private company?
Modella AI is a private company. It is classified as corporate owned and is currently acquired.
When was Modella AI founded?
Modella AI was founded in 2024. It employs 1 to 10 people.
Where is Modella AI based?
Modella AI is headquartered in Boston, United States, in the North America region.
How does Modella AI make money?
Three revenue lines are on record. AI Platform Licensing & Research Collaboration is the primary driver. The others are technology Licensing (Pre-Acquisition) and strategic Partnership Revenue.
Who are Modella AI's main competitors?
Direct peers on record are Lunit, Paige AI, Proscia, PathAI, Ibex Medical Analytics, Techcyte and Owkin. Tempus Labs is listed as a broad incumbent.
Does Modella AI have an API?
No public API is recorded for Modella AI.
What industry is Modella AI in?
Modella AI's product category is Computational Pathology AI. Its primary akta.pro industry code is HDAAAEAH, Medical Imaging AI, with a secondary code of HLAAAIAE, AI/ML-Enabled Biologic Discovery Platforms (computational design, in silico screening). Its NAICS code is 325413 and its SIC code is 3841.