Benchmark for LLMs' Intelligence in logic deduction/inference

Reasoning Capability benchmark for Meta.ai, ChatGPT and Gemini: score for 10-question reasoning:

 Click "Tests Score" for metrics :

See their collating Q&A table Answers Collation , which are collated from these individual Q&A sessions:

Thinking Test Cases for MetaAI.pdf,  Thinking Test Cases for ChatGPT.pdf,  Thinking Test Cases for Gemini.pdf

They contain spontaneous conversions between author and chatBots, include review of mistakes/pitfalls, corrections and summaries of learned so far

Contact Us

Santa Clara, CA, USA