Biomni

What it does

Biomni is a general-purpose biomedical AI agent for researchers. Users submit tasks in natural language—for example, variant interpretation, single-cell RNA-seq annotation, ADMET prediction, or CRISPR screen planning. The agent uses language-model reasoning, retrieval-augmented planning, and code execution across an environment described as containing specialized tools, databases, and software; it can return task results and generate hypotheses. The repository also documents saving execution history as a PDF.

The open-source software can be installed and run locally from Python, with an optional Gradio interface. This listing covers the open-source agent; the separate hosted app linked by the project has its own current access terms and feature set. Local setup requires a compatible LLM provider API key or a configured model endpoint; first-run data-lake downloads can be about 11 GB. The project warns that the agent executes generated code with full system privileges and recommends isolated or sandboxed environments. Integrated tools and data may have licenses more restrictive than Biomni’s Apache 2.0 license.

What it can do

• Plans and executes biomedical research tasks from natural-language instructions
• Interprets genetic variants
• Analyzes and annotates single-cell RNA-seq data
• Predicts ADMET properties
• Designs CRISPR screens and cloning protocols
• Saves agent execution history as PDF

Works with

• Anthropic
• OpenAI
• Azure OpenAI
• Google Gemini
• Groq
• AWS Bedrock
• Ollama
• Custom OpenAI-compatible model endpoints
• MCP servers

Before you choose

• The repository warns that LLM-generated code runs with full system privileges and may access files, the network, and system commands; use an isolated or sandboxed environment.
• The default data-lake download is about 11 GB; it can be skipped, but some tools may then be unavailable.
• Biomni is Apache 2.0 licensed, but integrated tools, databases, and software may have more restrictive licenses; the documented repository release was frozen in April 2025 and differs from the current web platform.

Listed under

• Data
• Research
• AI agent
• Bioinformatics
• Biomedical
• Clinical decision support
• Drug discovery
• Genomics

Get started

Follow repository setup and provider instructions.

1. Set up the environment using the repository’s setup instructions, then install Biomni with pip.
2. Configure credentials for a supported model provider or configure a custom model endpoint.
3. Initialize the agent and submit a biomedical task; allow for the first-run data-lake download unless it is skipped.

Apache-2.0 agent code; model-provider and compute costs are separate. Integrated tools and datasets have their own licenses.

Source: github.com/snap-stanford/biomni, checked 1 Nov 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 1 Nov 2026. Each source supports part of the overview; availability and prices can change.

• GitHub - snap-stanford/Biomni: Biomni: a general-purpose biomedical AI agent · GitHub ↗github.com
• Biomni - A General-Purpose Biomedical AI Agent ↗biomni.stanford.edu

Company website

github.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 1 Nov 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.