Is GPT-4 (0613) more capable than GPT-4 (0314)?
5
53
130
Jul 1
71%
chance

Resolve using the average over the four best benchmark comparisons, published in academic papers or preprint, at the time of resolution (e.g. four benchmarks: MATH, MMLU, HumanEval, SWAG).

Get Ṁ200 play money