Compare AI models

Specifications
Parameters
671B24B
Context window
128k32k
API pricing
Cheapest input
—$0.04DeepInfra
Cheapest output
—$0.064DeepInfra
—$14.00DeepInfra · fp8
Benchmarks
MMLU-Pro
84%66.3%
Overview
CompanyDeepSeekMistral
Release dateJan 20 2025Jan 30 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek R1 leads Mistral Small 3 on 1 of the 1 benchmark they both report (MMLU-Pro). DeepSeek R1 shipped 10 days before Mistral Small 3, so benchmark comparisons should account for the intervening progress.

DeepSeek R1 has 671B parameters, while Mistral Small 3 has 24B. Context windows are 128k (DeepSeek R1) vs 32k (Mistral Small 3).

On MMLU-Pro, DeepSeek R1 leads at 84% vs Mistral Small 3 at 66.3%.