← All comparisons

Phi-4 Multimodal Instruct vs Claude 4.1 Opus (Reasoning)

Microsoft vs Anthropic — side-by-side benchmark comparison

Phi-4 Multimodal InstructClaude 4.1 Opus (Reasoning)
Intelligence Index10.042.0
Coding Index36.5
Math Index80.3
Output speed (tok/s)16.644.5
Blended price ($/1M)$0.00$32.81
Time to first token (s)1.33s8.55s
aime9.3%
aime 2580.3%
artificial analysis coding index36.50
artificial analysis intelligence index10.0042.00
artificial analysis math index80.30
gpqa31.5%80.9%
hle4.4%11.9%
ifbench55.4%
lcr66.3%
livecodebench13.1%65.4%
math 50069.3%
mmlu pro48.5%88.0%
scicode11.0%40.9%
tau271.4%
terminalbench hard34.3%

Benchmark data from Artificial Analysis.