← All comparisons

Granite 4.0 H Small vs Claude Opus 4.6 (Adaptive Reasoning, Max Effort)

IBM vs Anthropic — side-by-side benchmark comparison

Granite 4.0 H SmallClaude Opus 4.6 (Adaptive Reasoning, Max Effort)
Intelligence Index10.852.9
Coding Index8.548.1
Math Index13.7
Output speed (tok/s)401.254.8
Blended price ($/1M)$0.11$10.94
Time to first token (s)8.75s11.69s
aime
aime 2513.7%
artificial analysis coding index8.5048.10
artificial analysis intelligence index10.8052.90
artificial analysis math index13.70
gpqa41.6%89.6%
hle3.7%36.7%
ifbench31.5%53.1%
lcr9.0%70.7%
livecodebench25.1%
math 500
mmlu pro62.4%
scicode20.9%51.9%
tau217.3%92.1%
terminalbench hard2.3%46.2%

Benchmark data from Artificial Analysis.