scientific-discovery/LLEMA
[ICLR 2026] LLEMA: Evolutionary Search with LLMs for Multi-Objective Materials Discovery
This project helps materials scientists and researchers discover new materials with specific properties faster. It takes your desired material characteristics (like high conductivity and low cost) and generates novel material compositions and structures. The output includes candidate materials, their predicted properties, and an evaluation of their stability and synthesizability.
Use this if you need to accelerate the discovery of new materials by intelligently exploring a vast chemical space and optimizing for multiple, potentially conflicting, properties.
Not ideal if you are looking for an off-the-shelf solution for existing material property prediction or if you don't have access to an OpenAI API key or Materials Project API key.
Stars
12
Forks
3
Language
Python
License
MIT
Category
Last pushed
Mar 09, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/scientific-discovery/LLEMA"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
CLAIRE-Labo/EvoTune
Efficiently discovering algorithms via LLMs with evolutionary search and reinforcement learning.
flowersteam/LLM-Culture
Code for the "Cultural evolution in populations of Large Language Models" paper
OpenMLRL/LLM_Collab_Code_Generation
LLM Collaboration for Code Generation
papachristoumarios/llm-network-formation
Supplementary Code and Data for "Network Formation and Dynamics among Multi-LLMs"
chirindaopensource/bias_adjusted_LLM_agents_human_like_decision_making
End-to-End Python framework implementing bias-adjusted LLM agents for human-like decision-making...