Back

Mistral

Magistral Small 1.2

ReasoningOpen weightsModels.devHugging Face
Intelligence
8.6
Rank 485 of 744 measured models
Coding
14.7
Rank 247 of 298 measured models
Output speed
74 t/s
Rank 241 of 403 measured models
Price/1M
$0.75
Rank 335 of 659 measured models
Context window
128K
Rank 773 of 1199 measured models

Magistral Small 1.2 scores 8.6 on the Artificial Analysis Intelligence Index, ranking 485th of 744 measured models. The closest models on this index are Claude 3 Opus (8.7), Devstral Small (May '25) (8.7), and Devstral 2 (8.6). At $0.75 per million tokens (3:1 input/output blended price), it is cheaper than Claude 3 Opus ($30). With an output speed of 74 t/s, it is slower than Devstral 2 (139 t/s).

Artificial Analysis Indices
Intelligence8.6
Rank 485 of 744 measured models
Coding14.7
Rank 247 of 298 measured models
Math80.3
Rank 64 of 270 measured models
Standard Benchmarks
MMLU Pro76.8%
Rank 158 of 349 measured models
GPQA Diamond66.3%
Rank 360 of 619 measured models
HLE6.4%
Rank 407 of 664 measured models
LiveCodeBench72.3%
Rank 60 of 343 measured models
AIME 202580.3%
Rank 64 of 270 measured models
IFBench44.4%
Rank 225 of 450 measured models
AA-LCR19.3%
Rank 455 of 575 measured models
TerminalBench Hard4.5%
Rank 299 of 432 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy