← All comparisons

GPT-5.4 mini (xhigh) vs Claude 3.5 Sonnet (Oct '24)

OpenAI vs Anthropic — side-by-side benchmark comparison

GPT-5.4 mini (xhigh)Claude 3.5 Sonnet (Oct '24)
Intelligence Index48.915.9
Coding Index51.530.2
Math Index
Output speed (tok/s)182.80.0
Blended price ($/1M)$1.69$6.56
Time to first token (s)4.25s0.00s
aime15.7%
aime 25
artificial analysis coding index51.5030.20
artificial analysis intelligence index48.9015.90
artificial analysis math index
gpqa87.5%59.9%
hle26.6%3.9%
ifbench73.3%
lcr69.3%
livecodebench38.1%
math 50077.1%
mmlu pro77.2%
scicode49.9%36.6%
tau283.3%
terminalbench hard52.3%

Benchmark data from Artificial Analysis.