Ai For Science Python Packages

tooluniverse

Democratizing AI scientists with ToolUniverse

agentsociety

AgentSociety 2 is a modern, LLM-native agent simulation platform designed for social science research and experimental design. It provides a flexible framework for creating and managing intelligent agents in simulated environments.

agentsociety2

AgentSociety 2 is a modern, LLM-native agent simulation platform designed for social science research and experimental design. It provides a flexible framework for creating and managing intelligent agents in simulated environments.

pytorchfire

GPU-accelerated wildfire simulator with differentiable cellular automata in PyTorch, supporting real-time parameter calibration via gradient descent

paperci

Continuous integration for scientific stories: traceable evidence, claim linting, and model-agnostic story design.

retrocast

RetroCast: A Unified Format for Multistep Retrosynthesis

ml-peg

ML Performance and Extrapolation Guide (reference deployment https://ml-peg.stfc.ac.uk)

psrn

[𝐍𝐚𝐭𝐮𝐫𝐞 𝐂𝐨𝐦𝐩𝐮𝐭𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐒𝐜𝐢𝐞𝐧𝐜𝐞] ⚡️ PSE/PSRN: Fast and efficient symbolic expression discovery through parallelized symbolic enumeration. Evaluates millions of expressions simultaneously on GPU with automated subtree reuse.

methods-mcp

Lightweight MCP server for paper methods extraction + reproducibility heuristics. Built for the Worldwide AI Science Fellowship.

hypothesaes

HypotheSAEs: hypothesizing interpretable relationships in text datasets using sparse autoencoders. https://arxiv.org/abs/2502.04382

curie-ai

❓Curie: Automated and Rigorous Scientific Experimentation with AI Agents

openaaas-mcp-adapter

OpenAaaS: science agent network — bring AI to your data, not your data to AI. AaaS, Agent as a Service, MCP protocol, local execution, Docker sandbox, zero-config Rust nodes.

scholaraio

Scholar All-In-One: A research infrastructure for AI agents

xscientist

Turn ideas into autonomous research with Git-like evidence histories—inspectable, reproducible, and reversible.

openmix

Open-source framework for computational formulation science from ingredient validation to autonomous mixture optimization. The missing infrastructure layer between single-molecule tools (RDKit) and real-world formulation.

preprint-fulltext

Retrieve full text of bioRxiv, medRxiv & arXiv preprints as structured sections — to ground LLM/agent scientific reasoning, deep research, and RAG. CLI + Python + MCP.

pyopenaaas

OpenAaaS: science agent network — bring AI to your data, not your data to AI. AaaS, Agent as a Service, MCP protocol, local execution, Docker sandbox, zero-config Rust nodes.

ppmat

Open-source AI infrastructure for materials science

oneehr

[KDD 2026] An all-in-one EHR predictive modeling and analysis library in Python

molecode

Molecode presents molecules as code and enables LLMs to operate and reason on chemistry directly.

popper-agent

Automated Hypothesis Testing with Agentic Sequential Falsifications

ppcfd

PaddleCFD is a deep learning toolkit for surrogate modeling, equation discovery, shape optimization and flow-control strategy discovery in the field of fluid mechanics.