Claude Science
Claude Science is Anthropic's beta AI workbench for scientists. It brings literature analysis, scientific artifacts, code-backed figures, domain skills, and compute orchestration into a Claude session that can run locally on macOS or Linux, over SSH, or through an HPC login node. The surface is best treated as a hybrid research workbench rather than a coding IDE: a coordinating agent can call curated science skills and specialist agents, while a reviewer agent checks citations, calculations, and whether figures trace back to their code.
Overview
Beta AI workbench for scientists combining literature analysis, scientific artifacts, curated science skills, auditable outputs, and flexible compute access on macOS, Linux, or remote SSH/HPC nodes.
Compare Claude Science side-by-side with other CLI, IDE, cloud, and framework agents on pricing, model coverage, and release cadence.
Best LLMs for this agent
Use the current Opus-class Claude when the session needs stronger multi-step reasoning across papers, datasets, code, and artifact review. Anthropic does not name a fixed model for Claude Science in the launch post, so this is an LLM Reference recommendation rather than a vendor-stated default.
Use the current Sonnet-class Claude when cost and latency matter more than maximum reasoning depth. Anthropic states Claude Science is available to Pro, Max, Team, and Enterprise users, but does not state a fixed backing model in the launch post.
Pricing and Provider Comparison
Hosted agents expose plan pricing. Bring-your-own-key agents pass the main cost through the model provider route you configure.
Featured Model Specs
Model Compatibility Matrix
Underlying Models
Compare to Similar Harnesses
FAQ
What is Claude Science?
Claude Science is Anthropic's AI workbench for scientists. It combines a Claude session with scientific tools, curated skills, auditable artifacts, and compute access for research workflows such as literature analysis, figure generation, genomics, protein structure work, and cheminformatics.
Where does Claude Science run?
Anthropic says Claude Science is available on macOS and Linux, and can also connect to a remote machine over SSH or an HPC login node. That makes it a hybrid surface: users keep work on their own machines or lab infrastructure while Claude receives the context needed for each step.
Does Claude Science use multiple agents?
Yes. Anthropic describes a generalist coordinating agent with more than 60 curated skills and connectors, the ability to spin up specialist agents, and a reviewer agent that checks citations, calculations, and whether figures match the underlying code.
Who can access Claude Science?
Anthropic released Claude Science in beta for Claude Pro, Max, Team, and Enterprise users. Team and Enterprise customers need an admin to enable the app.
Is Claude Science only for coding?
No. Claude Science includes code-backed workflows, but the product is positioned around scientific research: literature analysis, reproducible scientific artifacts, manuscript and figure iteration, domain databases, and compute orchestration.