← All comparisons

Claude 4.5 Haiku (Reasoning) vs Grok 3 mini Reasoning (high)

Anthropic vs xAI — side-by-side benchmark comparison

Claude 4.5 Haiku (Reasoning)Grok 3 mini Reasoning (high)
Intelligence Index37.132.1
Coding Index32.625.2
Math Index83.784.7
Output speed (tok/s)142.256.8
Blended price ($/1M)$2.19$0.35
Time to first token (s)10.48s0.42s
aime93.3%
aime 2583.7%84.7%
artificial analysis coding index32.6025.20
artificial analysis intelligence index37.1032.10
artificial analysis math index83.7084.70
gpqa67.2%79.1%
hle9.7%11.1%
ifbench54.3%45.9%
lcr70.3%50.3%
livecodebench61.5%69.6%
math 50099.2%
mmlu pro76.0%82.8%
scicode43.3%40.6%
tau254.7%90.4%
terminalbench hard27.3%17.4%

Benchmark data from Artificial Analysis.