1
Deep Intelligent Pharma
An AI-native, multi-agent platform transforming pharmaceutical R&D by reimagining discovery and development with autonomous agents and intelligent databases.
AI Tool
Co-Scientist Review
Co-Scientist is a multi-agent AI partner built on Gemini, designed to accelerate scientific research by generating and refining hypotheses.
shipped Jun 3, 2026aifreemium
1Formally introduced in a Nature paper on May 19, 2026, by Google DeepMind.
2Built on the Gemini model, utilizing a multi-agent architecture with specialized agents like Proximity, Reflection, and Ranking.
3Developed in collaboration with researchers from over 100 institutions globally.
4Demonstrated ability to recreate a decade of antibiotic resistance research in 72 hours.
Co-Scientist at a Glance
About Co-Scientist
Leadership
Demis HassabisCo-founder & CEO
Shane LeggCo-founder
Mustafa SuleymanCo-founder
Similar Tools
Compare Alternatives
Other tools you might consider
Visit→
2
FutureHouse
An AI platform designed to automate critical steps in scientific research, utilizing a multi-agent workflow for tasks like literature searches, data analysis, and hypothesis generation.
3
BenevolentAI
Applies machine learning to biomedical research to identify connections in vast datasets and generate novel hypotheses for drug discovery.
4
HyperWrite (Hypothesis Maker)
An AI-driven tool that generates a hypothesis based on a user's research question, leveraging advanced AI models.
Connect
𝕏
X / Twitter@GoogleDeepMind
overview
What is Co-Scientist?
Co-Scientist is a multi-agent AI system developed by Google DeepMind that enables researchers and scientists to generate and evolve hypotheses to accelerate scientific breakthroughs. It was formally introduced in a Nature paper on May 19, 2026, and is built on the Gemini model. This system functions as a "scientist-in-the-loop" intelligent assistant, augmenting human expertise by generating, debating, and evolving novel hypotheses for complex scientific problems. Its architecture includes specialized agents such as Proximity for mapping and clustering hypotheses, Reflection for critiquing correctness and novelty, Ranking for scientific debates, and Evolution for refining ideas, all orchestrated by a Supervisor agent. Co-Scientist is being rolled out to individual researchers through an experimental tool called Hypothesis Generation, jointly developed across Google DeepMind, Google Research, Google Cloud, and Google Labs, as part of the broader "Gemini for Science" initiative.
quick facts
Quick Facts
| Attribute | Value |
|---|---|
| Developer | Google DeepMind |
| Business Model | Freemium |
| Pricing | Freemium |
| Platforms | Web |
| API Available | No |
| Founded | 2010 (DeepMind) |
| HQ | Mountain View, USA |
| Funding | Acquired (by Google) |
features
Key Features of Co-Scientist
Co-Scientist leverages its multi-agent AI architecture to provide a suite of capabilities designed to enhance scientific research and discovery. These features are built upon the Gemini model and aim to streamline various stages of the research process, from initial ideation to experimental proposal.
- 1Multi-agent AI system built on Gemini technology for complex problem-solving.
- 2Novel hypothesis generation based on natural language research goals.
- 3Hypothesis refinement and evolution through internal 'scientific debates' among agents.
- 4Integration of web search and specialized databases (e.g., ChEMBL, UniProt) for literature synthesis.
- 5Capability to leverage advanced models like AlphaFold in select collaborations.
- 6Assistance in optimizing experimental design and analyzing complex datasets.
- 7Identification of novel treatment targets and drug repurposing candidates.
- 8Uncovering original knowledge and formulating experimental proposals.
- 9Accelerating genetic research by synthesizing literature and analyzing large screening datasets.
use cases
Who Should Use Co-Scientist?
Co-Scientist is primarily designed for researchers and scientists across various disciplines who seek to accelerate their discovery processes, generate novel hypotheses, and synthesize complex scientific literature more efficiently. Its capabilities are particularly beneficial in fields requiring extensive data analysis and the formulation of new research directions.
- 1**Researchers and Scientists:** For accelerating scientific breakthroughs across various disciplines, including biology, chemistry, and medicine.
- 2**Drug Discovery Teams:** For novel drug repurposing, identifying new treatment targets, and understanding disease mechanisms.
- 3**Geneticists and Biologists:** For speeding up research on cellular processes, such as reversing cellular aging, by synthesizing literature and analyzing large screening datasets.
- 4**Academics and Industrial R&D:** For literature review, evidence synthesis, and formulating novel research hypotheses and experimental proposals.
pricing
Co-Scientist Pricing & Plans
Co-Scientist operates on a freemium model, indicating that a basic version of the tool is available at no cost, with potential for premium features or expanded access through paid plans. Specific details regarding the tiers, pricing structure, or what constitutes the premium offerings have not been publicly disclosed beyond the freemium designation.
- 1Free tier available (specific features not detailed).
- 2Premium plans with advanced features and expanded access (details not publicly disclosed).
competitors
Co-Scientist vs Competitors
Co-Scientist operates within a growing landscape of AI tools aimed at accelerating scientific research. Its multi-agent architecture and direct integration with Google's Gemini model provide distinct advantages and differentiators when compared to other platforms.
1
Deep Intelligent Pharma↗
An AI-native, multi-agent platform transforming pharmaceutical R&D by reimagining discovery and development with autonomous agents and intelligent databases.
Similar to Co-Scientist in its multi-agent approach and focus on accelerating discovery, but specifically targets pharmaceutical R&D, whereas Co-Scientist appears more general across scientific breakthroughs. Pricing information is not readily available, suggesting a different model than Co-Scientist's freemium.
2
FutureHouse↗
An AI platform designed to automate critical steps in scientific research, utilizing a multi-agent workflow for tasks like literature searches, data analysis, and hypothesis generation.
Directly comparable to Co-Scientist in its broad goal of accelerating scientific progress through AI and multi-agent systems, covering various research stages including hypothesis generation. Pricing is not explicitly stated as freemium, which might differentiate it from Co-Scientist.
3
BenevolentAI↗
Applies machine learning to biomedical research to identify connections in vast datasets and generate novel hypotheses for drug discovery.
While Co-Scientist is a general multi-agent AI for hypothesis generation, BenevolentAI is specifically focused on drug discovery within biomedical research, using machine learning rather than explicitly a multi-agent system for evolving hypotheses. Its target audience is pharmaceutical companies, differing from Co-Scientist's broader researcher audience.
4
An AI-driven tool that generates a hypothesis based on a user's research question, leveraging advanced AI models.
HyperWrite's Hypothesis Maker offers a direct hypothesis generation feature, similar to a core function of Co-Scientist, but it is a simpler tool focused on single-shot generation rather than a multi-agent system for evolving hypotheses. It offers a limited free trial with premium plans, aligning with Co-Scientist's freemium model.
❓
Frequently Asked Questions
+What is Co-Scientist?
Co-Scientist is a multi-agent AI system developed by Google DeepMind that enables researchers and scientists to generate and evolve hypotheses to accelerate scientific breakthroughs. It was formally introduced in a Nature paper on May 19, 2026, and is built on the Gemini model.
+Is Co-Scientist free?
Co-Scientist operates on a freemium model, meaning a basic version is available for free. Details regarding specific premium plans or advanced feature costs have not been publicly disclosed.
+What are the main features of Co-Scientist?
Key features include multi-agent AI for hypothesis generation and refinement, integration with specialized databases for literature synthesis, assistance in experimental design, and capabilities for drug repurposing and target discovery. It is built on Google's Gemini technology.
+Who should use Co-Scientist?
Co-Scientist is intended for researchers and scientists across various disciplines, including those in drug discovery, genetics, and academic or industrial R&D, who aim to accelerate scientific discovery and generate novel hypotheses.
+How does Co-Scientist compare to alternatives?
Co-Scientist differentiates itself through its multi-agent architecture built on Gemini, offering comprehensive hypothesis evolution and broad scientific application. Competitors like Deep Intelligent Pharma and BenevolentAI often focus more narrowly on pharmaceutical R&D, while tools like HyperWrite's Hypothesis Maker provide simpler, single-shot hypothesis generation without the multi-agent collaborative framework.
More on Stork
Related AI Tools
Other tools in this category, ranked by community signal
Incogni
🤖 AI Tools
Incogni helps users remove their personal data from data brokers who collect, aggregate, and trade this information without consent.
PandaProbe Cloud
🤖 AI Tools
PandaProbe Cloud offers production-grade agent tracing, evaluations, and monitoring services that are fully managed, eliminating infrastructure overhead for teams.
Flightmare
🤖 AI Tools
Flightmare is a tool designed for simulating and testing flight scenarios. It provides a comprehensive documentation for users to understand its functionalities and usage.
Serve Robotics (sidewalk robots)
🤖 AI Tools
Serve Robotics focuses on sustainable, self-driving delivery solutions, aiming to optimize the delivery of small items like burritos using autonomous vehicles.
docAnalyzer
🤖 AI Tools
A source-grounded AI workspace for expert practitioners. Three chat modes, dozens of models in one plan, and downloadable files that compound across turns.
Bugpilot
🤖 AI Tools
Capture console, network & DOM as AI-ready Markdown for instant bug reports. Built for Claude, ChatGPT & AI coding tools.
For builders
This page is doing a job for someone else’s tool.
AI agents read it. Buyers find it. Backlinks accrue. Your tool can have one too — live in 24 hours, indexed by Claude, ChatGPT, and Perplexity, queryable via MCP.