Compare AI models

Specifications
Parameters
671B—
Context window
128k128k
API pricing
Cheapest input
$0.25SiliconFlow—
Cheapest output
$1.00SiliconFlow—
$100SiliconFlow · fp8—
Benchmarks
LiveCodeBench
49.2%75%
AIME 2024
59.4%91.82%
GPQA Diamond
68.4%76.26%
Overview
CompanyDeepSeekMistral
Release dateMar 24 2025Sep 17 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Magistral Medium 1.2 leads DeepSeek V3-0324 on 3 of the 3 benchmarks they both report (LiveCodeBench, AIME 2024, GPQA Diamond). DeepSeek V3-0324 shipped 177 days before Magistral Medium 1.2, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3-0324) vs 128k (Magistral Medium 1.2). DeepSeek V3-0324 is open weight, while Magistral Medium 1.2 is closed.

On LiveCodeBench, Magistral Medium 1.2 leads at 75% vs DeepSeek V3-0324 at 49.2%. On AIME 2024, Magistral Medium 1.2 leads at 91.82% vs DeepSeek V3-0324 at 59.4%. On GPQA Diamond, Magistral Medium 1.2 leads at 76.26% vs DeepSeek V3-0324 at 68.4%.