← All comparisons

GPT-5.2 (xhigh) vs Claude 3.5 Sonnet (June '24)

OpenAI vs Anthropic — side-by-side benchmark comparison

GPT-5.2 (xhigh)Claude 3.5 Sonnet (June '24)
Intelligence Index51.314.2
Coding Index48.726.0
Math Index99.0
Output speed (tok/s)77.20.0
Blended price ($/1M)$4.81$6.56
Time to first token (s)64.17s0.00s
aime9.7%
aime 2599.0%
artificial analysis coding index48.7026.00
artificial analysis intelligence index51.3014.20
artificial analysis math index99.00
gpqa90.3%56.0%
hle35.4%3.7%
ifbench75.4%
lcr72.7%
livecodebench88.9%
math 50069.5%
mmlu pro87.4%75.1%
scicode52.1%31.6%
tau284.8%
terminalbench hard47.0%

Benchmark data from Artificial Analysis.