Tahoe Therapeutics

What Tahoe Therapeutics does

Tahoe Therapeutics is a privately held biotechnology company founded in 2022 (originally as Vevo Therapeutics, rebranded April 2025) and headquartered in South San Francisco, California. The company develops AI-powered virtual cell models to accelerate drug discovery, with an initial focus on oncology. Its core offering centers on the proprietary Mosaic platform, which generates gigascale perturbative single-cell data measuring how drug molecules interact with patient biology across hundreds of cancer cell lines. The flagship Tahoe-100M dataset, open-sourced in February 2025, contains 100 million cells mapping 60,000 drug-patient interactions across 50 cancer cell lines and 1,200 drug perturbations — described as 50x larger than all public perturbative single-cell data combined. Building on this data layer, Tahoe has released Tahoe-x1, a 3-billion-parameter open-source foundation model that achieves state-of-the-art performance on gene essentiality prediction, plus TahoeDive (a natural-language AI agent) and Rhaister (a lightweight statistical model for drug phenotype prediction).

Tahoe operates a dual business model. On the platform side, it selectively partners with pharmaceutical and biotech companies to share forthcoming billion-cell datasets under negotiated licensing and co-development terms. It also pursues a co-investment model via joint ventures, exemplified by the January 2026 joint venture with Alloy Therapeutics to develop first-in-class antibody-drug conjugates (ADCs) for hard-to-treat cancers. On the proprietary side, the company is advancing its own therapeutic programs toward the clinic, supported by a drug discovery organization that includes veterans from Plexxikon behind the FDA-approved drugs vemurafenib and pexidartinib. The go-to-market combines enterprise partnership sales, joint venture formation, and an open-source community strategy that has driven 250,000+ downloads of Tahoe-100M and selection into the AWS Generative AI Accelerator (2023) and the Databricks Generative AI Startup Challenge (2024).

The company has raised approximately $42 million across two rounds: a $12M seed in December 2022 co-led by Wing Venture Capital and General Catalyst, and a $30M Series A in August 2025 led by Amplify Partners with participation from Databricks Ventures and other institutional investors. The team of approximately 20-30 is led by four co-founders — CEO Nima Alidoust (ex-Rigetti, McKinsey, 1QBit), CSO Johnny Yu (ex-Biogen, Broad Institute, Millennium Pharmaceuticals), Hani Goodarzi (UCSF Associate Professor, Mosaic platform inventor), and Kevan Shokat (UCSF and UC Berkeley Professor, HHMI investigator) — and recently added VP-level drug discovery leadership to operationalize its therapeutic pipeline.

Name

Tahoe Therapeutics

Legal name

Tahoe Bio

Website

https://tahoebio.ai

Company type

Private

Founded year

2022

Operating status

Operating

Headcount range

11–50 employees

Short description

Tahoe Therapeutics builds AI-powered virtual cell models and gigascale perturbative single-cell datasets (Tahoe-100M) to accelerate oncology drug discovery, serving pharmaceutical partners via data licensing and joint ventures while advancing its own therapeutic pipeline.

Ownership category

akta.pro rank

Product category

AI Drug Discovery / Computational Biology

NAICS

Research and Development in Biotechnology (except Nanobiotechnology) (541714)

SIC

Services-Commercial Physical & Biological Research (8731)

akta.pro primary industry

AI/ML-Enabled Biologic Discovery Platforms (computational design, in silico screening) (HLAAAIAE)

akta.pro secondary industries

Drug Discovery (Target ID/Validation & Hit Discovery) (HLAIALAA), Single-Cell & Spatial Omics Discovery Platforms (target ID/validation, functional profiling) (HLAAAIAF)

Keywords

• AI drug discovery
• Virtual cell models
• Single-cell perturbation data
• Computational biology foundation models
• Precision medicine oncology

Headquarters

HQ city

South San Francisco

HQ country

United States

HQ region

North America

Offices2 records

Markets served

GTM type

B2B

Offering type

Software

Cost components

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

Revenue model

1. Strategic Data Partnerships: Tahoe selects single pharmaceutical or AI partners to access its forthcoming billion-cell dataset. The partner gains access to data combined with Tahoe's clinical or modeling expertise to develop medicines powered by virtual cell models. This represents potential licensing/access fees and co-development economics.
2. Proprietary Therapeutic Programs: Company is advancing its own therapeutic programs toward the clinic. Revenue would come from successful drug development through clinical trials and eventual commercialization, though programs are still early-stage.
3. Joint Venture Revenue Share: Tahoe and Alloy Therapeutics co-invest and co-build a new ADC development company. Tahoe contributes novel targets and biomarker insights. The joint venture aims to advance ADC programs to value-inflection points suitable for independent financing or pharma partnerships, with revenue sharing between partners.

Pricing tiers

Go-to-market motion3 records

Distribution channels4 records

Marketing channels10 records

Core offering

Tahoe Therapeutics develops AI-powered virtual cell models and gigascale perturbative single-cell datasets for drug discovery, beginning in oncology. Its proprietary Mosaic platform generates large-scale in vivo drug response data at single-cell precision, which is used to train foundation models (e.g., Tahoe-x1) that predict cellular responses and identify novel therapeutic targets. The company monetizes through strategic data partnerships with a single pharmaceutical partner, joint ventures for therapeutic co-development (such as the Alloy Therapeutics ADC JV), and proprietary in-house drug programs.

Product overview

Tahoe Therapeutics (formerly Vevo Therapeutics) offers an AI-powered drug discovery platform centered on its proprietary Mosaic technology, which generates gigascale perturbative single-cell datasets. The core product portfolio includes: (1) Mosaic Platform - the foundational technology for scalable in vivo single-cell drug response measurement; (2) Tahoe-100M - the world's largest open-source single-cell perturbation dataset (100M cells, 60,000 drug-patient interactions); (3) TahoeDive - a natural language AI agent for querying the dataset; (4) Tahoe-x1 - a 3-billion-parameter foundation model for cell biology; (5) Rhaister - a minimalist statistical model for drug phenotype prediction; and (6) Emerald Bay - a 2M cell sensitivity dataset. The platform enables AI-driven virtual cell models for drug discovery and precision medicine, with a focus on oncology.

Differentiator

Problem solved

Functional benefit

Brands

• Tahoe-x1: A 3-billion-parameter foundation model that learns unified representations of genes, cells, and drugs, achieving state-of-the-art performance across cancer-relevant single-cell biology benchmarks.
• TahoeDive
• Rhaister
• Emerald Bay
• Mosaic Platform

Products and services

• Mosaic Platform Proprietary technology platform that makes in vivo drug response data generation scalable with single-cell precision. Generates gigascale perturbative single-cell maps measuring how drug molecules interact with patient biology across hundreds of diverse patient samples using a cell-village approach that eliminates batch effects. Licensed exclusively from UCSF Innovation Ventures.
• Tahoe-100M World's largest single-cell perturbation dataset: 100 million cells mapping 60,000 drug-patient interactions across 50 cancer models and 1,200 drug perturbations. 50x larger than all public perturbative single-cell data combined. Open-sourced as the inaugural contribution to Arc Institute's Virtual Cell Atlas and downloaded 250,000+ times in less than a year.
• Tahoe-x1 (Tx1) 3-billion-parameter foundation model trained on perturbation-rich single-cell data. First billion-parameter, compute-efficient foundation model for cell-state prediction. Achieves state-of-the-art performance on DepMap gene essentiality prediction and MSigDB oncogenic program benchmarks. 3-30x more compute-efficient than previous cell-state models. Fully open-source with weights, training and evaluation code released.
• TahoeDive AI agent built in partnership with Kepler AI that empowers biologists to query and analyze the Tahoe-100M dataset using natural language while accounting for broader scientific literature. Generates code to analyze data and visualize results. Available in beta.
• Rhaister Minimalist statistical model that predicts drug phenotypes in new contexts with accuracies comparable to experimental assays. Trained in seconds and produces predictions in milliseconds. First model to show consistent scaling with more perturbation data and demonstrate significant zero-shot performance. Open-source.
• Emerald Bay Dataset 2 million cell dataset measuring long time-course (5-day) sensitivity phenotypes across thousands of drug-cell line interactions. Generated using the Mosaic platform. Open-sourced alongside Rhaister.

Quantifiable outcome

• 100M single-cell datapoints generated (Tahoe-100M)
• +5 more outcomes

Named customers1 record

Segments3 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability6 records

Feature6 records

Scale indicators10 records

Recent moves7 records

Expansion highlights5 records

Direct peers

• Recursion Pharmaceuticals: AI-driven drug discovery company combining high-throughput biology with machine learning at scale. Most directly comparable to Tahoe in using industrial-scale experimental data generation paired with foundation models to discover new therapeutics.
• Insitro: ML-enabled drug discovery company building data-generation platforms and predictive models for therapeutic discovery. Comparable to Tahoe's combination of proprietary wet-lab data generation with AI models for target and drug identification.
• Generate Biomedicines: Generative biology company building AI models to design novel protein therapeutics. Comparable in pursuing foundation-model approaches to biology and pursuing both partnership and proprietary drug strategies.
• Cellarity: Cell-behavior-based drug discovery company using AI to model cellular states. Most directly comparable to Tahoe's 'virtual cell' thesis — both companies build ML models of cell behavior for drug discovery rather than starting from target-based chemistry.

Broad incumbents

• Isomorphic Labs: Alphabet/DeepMind-backed drug discovery company building virtual cell / foundation models of biology with significantly greater capital. Operates in the same end-market as Tahoe and competes for pharma partnerships.
• Ginkgo Bioworks: Synthetic biology platform serving pharma and industrial biotech with cell-engineering and data-generation capabilities. Comparable as a large-scale biology platform with B2B partnership model and proprietary data assets, though focused on engineered organisms rather than virtual cell models.

Emerging players

• EvolutionaryScale: AI-frontier lab building biological foundation models (ESM3) for protein and cell-state understanding. Comparable to Tahoe in using large-scale training data and foundation models for biology, with overlap in target ID and virtual-cell aspirations.
• Cradle Bio: AI protein-design platform serving biotech and pharma partners. Comparable to Tahoe in the ML-for-biology category, with overlap in serving pharma customers seeking to accelerate biologics discovery.
• Genesis Therapeutics: AI-driven drug discovery company combining molecular ML with foundation models. Comparable in pursuing AI foundation models for therapeutic discovery with a hybrid platform-plus-pipeline model.
• Arc Institute: Research institute co-leading the Virtual Cell Atlas with Tahoe. Comparable in building open-source, perturbation-rich datasets and foundation models for cell biology; functions as both a collaborator and an adjacent potential competitor in the virtual-cell-model category.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks7 records

Key highlights7 records

Customer concentration

• Xx.com/tahoe_ai
• GitHubgithub.com/tahoebio
• LinkedInlinkedin.com/company/tahoe-therapeutics

Revenue estimate

Valuation estimate

Number of profiles

Profiles6 records

Subsidiaries1 record

Funding overview

Funding rounds2 records

Investors10 records

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 Tahoe Therapeutics