Claude 3 Opus vs GPT-4 (Mar 2023)

Verdict

Claude 3 Opus leads 4–1 across 5 shared benchmarks.

Claude 3 Opus 4 · GPT-4 (Mar 2023) 1 · higher isn't always better — check the caveats below.

Per-benchmark head-to-head

Reported scores — protocols may differ. How we compare

Sort

Claude 3 OpusGPT-4 (Mar 2023)

29.55
BGPQA diamond
14.31
37.75
AMETR Time Horizons
36.11
79.47
CMMLU
81.87
4.63
AOTIS Mock AIME 2024-2025
0.46
77
CWinogrande
75

A higher score does not always mean a better model. Each number is task performance under a disclosed harness. Of the 10 measurements across 5 shared benchmarks, 10 single-evaluator or undisclosed-protocol reports; 40% are independently reproduced. Check any benchmark's integrity grade before reading a win as decisive.

Frequently asked questions

Claude 3 Opus leads 4–1 across 5 shared benchmarks. Claude 3 Opus leads 4 benchmarks and GPT-4 (Mar 2023) leads 1. Higher isn't always better — see the integrity caveats on each benchmark.

Claude 3 Opus and GPT-4 (Mar 2023) have 5 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.

Claude 3 Opus vs GPT-4 (Mar 2023)

Per-benchmark head-to-head

Frequently asked questions

Follow the record

These models

Same tier

Claude 3 Opus vs GPT-4 (Mar 2023)

Per-benchmark head-to-head

Frequently asked questions

Follow the record

These models

Same tier