← All comparisons

Phi-4 Mini Instruct vs Claude 3.5 Sonnet (Oct '24)

Microsoft vs Anthropic — side-by-side benchmark comparison

Phi-4 Mini InstructClaude 3.5 Sonnet (Oct '24)
Intelligence Index8.415.9
Coding Index3.630.2
Math Index6.7
Output speed (tok/s)0.00.0
Blended price ($/1M)$0.00$6.56
Time to first token (s)0.00s0.00s
aime3.0%15.7%
aime 256.7%
artificial analysis coding index3.6030.20
artificial analysis intelligence index8.4015.90
artificial analysis math index6.70
gpqa33.1%59.9%
hle4.2%3.9%
ifbench21.1%
lcr13.7%
livecodebench12.6%38.1%
math 50069.6%77.1%
mmlu pro46.5%77.2%
scicode10.8%36.6%
tau28.2%
terminalbench hard0.0%

Benchmark data from Artificial Analysis.