← All comparisons

Magistral Small 1.2 vs Claude 4.1 Opus (Reasoning)

Mistral vs Anthropic — side-by-side benchmark comparison

Magistral Small 1.2Claude 4.1 Opus (Reasoning)
Intelligence Index18.242.0
Coding Index14.836.5
Math Index80.380.3
Output speed (tok/s)111.144.5
Blended price ($/1M)$0.75$32.81
Time to first token (s)0.38s8.55s
aime
aime 2580.3%80.3%
artificial analysis coding index14.8036.50
artificial analysis intelligence index18.2042.00
artificial analysis math index80.3080.30
gpqa66.3%80.9%
hle6.1%11.9%
ifbench44.4%55.4%
lcr16.3%66.3%
livecodebench72.3%65.4%
math 500
mmlu pro76.8%88.0%
scicode35.2%40.9%
tau227.8%71.4%
terminalbench hard4.5%34.3%

Benchmark data from Artificial Analysis.