Compare AI models

Specifications
Parameters
16B24B
Context window
32k32k
API pricing
Cheapest input
—$0.04DeepInfra
Cheapest output
—$0.064DeepInfra
—$14.00DeepInfra · fp8
Benchmarks
HumanEval
57.3%84.8%
MATH
27.9%70.6%
Overview
CompanyDeepSeekMistral
Release dateMay 16 2024Jan 30 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Small 3 leads DeepSeek V2 Lite on 2 of the 2 benchmarks they both report (HumanEval, MATH). DeepSeek V2 Lite shipped 259 days before Mistral Small 3, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Mistral Small 3 has 24B. Context windows are 32k (DeepSeek V2 Lite) vs 32k (Mistral Small 3).

On HumanEval, Mistral Small 3 leads at 84.8% vs DeepSeek V2 Lite at 57.3%. On MATH, Mistral Small 3 leads at 70.6% vs DeepSeek V2 Lite at 27.9%.