← All comparisons

Claude 4.5 Haiku (Non-reasoning) vs Sonar

Anthropic vs Perplexity — side-by-side benchmark comparison

Claude 4.5 Haiku (Non-reasoning)Sonar
Intelligence Index31.015.5
Coding Index29.6
Math Index39.0
Output speed (tok/s)105.80.0
Blended price ($/1M)$2.19$0.00
Time to first token (s)0.65s0.00s
aime48.7%
aime 2539.0%
artificial analysis coding index29.60
artificial analysis intelligence index31.0015.50
artificial analysis math index39.00
gpqa64.6%47.1%
hle4.3%7.3%
ifbench42.0%
lcr43.7%
livecodebench51.1%29.5%
math 50081.7%
mmlu pro80.0%68.9%
scicode34.4%22.9%
tau232.5%
terminalbench hard27.3%

Benchmark data from Artificial Analysis.