← All comparisons

Granite 4.1 8B vs Claude 3.5 Sonnet (Oct '24)

IBM vs Anthropic — side-by-side benchmark comparison

Granite 4.1 8BClaude 3.5 Sonnet (Oct '24)
Intelligence Index12.415.9
Coding Index7.330.2
Math Index
Output speed (tok/s)111.50.0
Blended price ($/1M)$0.06$6.56
Time to first token (s)0.41s0.00s
aime15.7%
aime 25
artificial analysis coding index7.3030.20
artificial analysis intelligence index12.4015.90
artificial analysis math index
gpqa43.3%59.9%
hle3.8%3.9%
ifbench38.6%
lcr12.0%
livecodebench38.1%
math 50077.1%
mmlu pro77.2%
scicode21.8%36.6%
tau227.8%
terminalbench hard0.0%

Benchmark data from Artificial Analysis.