← All comparisons

Magistral Medium 1.2 vs DeepSeek V3.1 Terminus (Reasoning)

Mistral vs DeepSeek — side-by-side benchmark comparison

Magistral Medium 1.2DeepSeek V3.1 Terminus (Reasoning)
Intelligence Index27.133.9
Coding Index21.733.7
Math Index82.089.7
Output speed (tok/s)39.40.0
Blended price ($/1M)$2.75$1.91
Time to first token (s)0.54s0.00s
aime
aime 2582.0%89.7%
artificial analysis coding index21.7033.70
artificial analysis intelligence index27.1033.90
artificial analysis math index82.0089.70
gpqa73.9%79.2%
hle9.6%15.2%
ifbench43.0%57.0%
lcr51.3%65.0%
livecodebench75.0%79.8%
math 500
mmlu pro81.5%85.1%
scicode39.2%40.6%
tau252.0%37.1%
terminalbench hard12.9%30.3%

Benchmark data from Artificial Analysis.