← All comparisons

Claude 4.5 Haiku (Reasoning) vs Apertus 70B Instruct

Anthropic vs Swiss AI Initiative — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)Apertus 70B Instruct
Intelligence Index37.17.7
Coding Index32.61.9
Math Index83.7
Output speed (tok/s)142.20.0
Blended price ($/1M)$2.19$1.34
Time to first token (s)10.48s0.00s
aime
aime 2583.7%
artificial analysis coding index32.601.90
artificial analysis intelligence index37.107.70
artificial analysis math index83.70
gpqa67.2%27.2%
hle9.7%5.5%
ifbench54.3%25.9%
lcr70.3%0.0%
livecodebench61.5%
math 500
mmlu pro76.0%
scicode43.3%5.7%
tau254.7%12.9%
terminalbench hard27.3%0.0%

Benchmark data from Artificial Analysis.