Compare AI models

Specifications
Parameters
141B8B
Context window
64k8k
API pricing
Cheapest input
$2.00Mistral—
Cheapest output
$6.00Mistral—
$700Mistral · zdr—
Benchmarks
HumanEval
45.1%36.22%
ARC-Challenge
91.3%64.42%
HellaSwag
88.5%83.03%
WinoGrande
84.7%80.35%
MMLU
77.75%69.51%
GSM8K
78.6%58.45%
Overview
CompanyMistralNVIDIA
Release dateApr 10 2024Aug 21 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mixtral 8×22B leads Mistral NeMo Minitron 8B on 6 of the 6 benchmarks they both report. Mixtral 8×22B shipped 133 days before Mistral NeMo Minitron 8B, so benchmark comparisons should account for the intervening progress.

Mixtral 8×22B has 141B parameters, while Mistral NeMo Minitron 8B has 8B. Context windows are 64k (Mixtral 8×22B) vs 8k (Mistral NeMo Minitron 8B).

On HumanEval, Mixtral 8×22B leads at 45.1% vs Mistral NeMo Minitron 8B at 36.22%. On ARC-Challenge, Mixtral 8×22B leads at 91.3% vs Mistral NeMo Minitron 8B at 64.42%. On HellaSwag, Mixtral 8×22B leads at 88.5% vs Mistral NeMo Minitron 8B at 83.03%. On WinoGrande, Mixtral 8×22B leads at 84.7% vs Mistral NeMo Minitron 8B at 80.35%. On MMLU, Mixtral 8×22B leads at 77.75% vs Mistral NeMo Minitron 8B at 69.51%. On GSM8K, Mixtral 8×22B leads at 78.6% vs Mistral NeMo Minitron 8B at 58.45%.