← All comparisons

Claude 4.5 Haiku (Non-reasoning) vs Grok 3

Anthropic vs xAI — side-by-side benchmark comparison

Claude 4.5 Haiku (Non-reasoning)Grok 3
Intelligence Index31.025.2
Coding Index29.619.8
Math Index39.058.0
Output speed (tok/s)105.80.0
Blended price ($/1M)$2.19$8.00
Time to first token (s)0.65s0.00s
aime33.0%
aime 2539.0%58.0%
artificial analysis coding index29.6019.80
artificial analysis intelligence index31.0025.20
artificial analysis math index39.0058.00
gpqa64.6%69.3%
hle4.3%5.1%
ifbench42.0%46.9%
lcr43.7%54.7%
livecodebench51.1%42.5%
math 50087.0%
mmlu pro80.0%79.9%
scicode34.4%36.8%
tau232.5%48.8%
terminalbench hard27.3%11.4%

Benchmark data from Artificial Analysis.