← All comparisons

Claude 4.5 Haiku (Reasoning) vs INTELLECT-3

Anthropic vs Prime Intellect — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)INTELLECT-3
Intelligence Index37.122.2
Coding Index32.619.1
Math Index83.788.0
Output speed (tok/s)142.20.0
Blended price ($/1M)$2.19$0.00
Time to first token (s)10.48s0.00s
aime
aime 2583.7%88.0%
artificial analysis coding index32.6019.10
artificial analysis intelligence index37.1022.20
artificial analysis math index83.7088.00
gpqa67.2%76.1%
hle9.7%12.1%
ifbench54.3%34.0%
lcr70.3%32.3%
livecodebench61.5%77.7%
math 500
mmlu pro76.0%82.2%
scicode43.3%39.1%
tau254.7%26.6%
terminalbench hard27.3%9.1%

Benchmark data from Artificial Analysis.