linhaowei1/kumo

☁️ KUMO: Generative Evaluation of Complex Reasoning in Large Language Models

29
/ 100
Experimental

This project helps evaluate how well large language models (LLMs) can solve complex reasoning problems. It takes a set of predefined truths, actions, and outcomes — like symptoms, tests, and diagnoses in medicine — and generates detailed reasoning games. The output is a benchmark that assesses the LLM's ability to deduce the correct truth efficiently. It's designed for AI researchers, machine learning engineers, and data scientists who are developing or comparing LLMs for tasks requiring logical deduction.

No commits in the last 6 months.

Use this if you need to rigorously test the complex reasoning capabilities of large language models across various domain-specific scenarios, using procedurally generated tasks.

Not ideal if you are looking for a simple, off-the-shelf evaluation of basic language understanding or generation tasks, or if you don't work with LLMs.

LLM evaluation AI research reasoning assessment machine learning engineering AI model benchmarking
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 5 / 25

How are scores calculated?

Stars

19

Forks

1

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Jun 04, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/linhaowei1/kumo"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.