Compare AI models

Specifications
Parameters
8B/70B/405B8B
Context window
128k8k
API pricing
Cheapest input
$0.02DeepInfra—
Cheapest output
$0.04DeepInfra—
$6.00DeepInfra · fp8—
Benchmarks
HumanEval
89%36.22%
ARC-Challenge
96.9%64.42%
MMLU
87.3%69.51%
GSM8K
96.8%58.45%
Overview
CompanyMetaNVIDIA
Release dateJul 23 2024Aug 21 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

LLaMA 3.1 leads Mistral NeMo Minitron 8B on 4 of the 4 benchmarks they both report (HumanEval, ARC-Challenge, MMLU, GSM8K). LLaMA 3.1 shipped 29 days before Mistral NeMo Minitron 8B, so benchmark comparisons should account for the intervening progress.

LLaMA 3.1 has 8B/70B/405B parameters, while Mistral NeMo Minitron 8B has 8B. Context windows are 128k (LLaMA 3.1) vs 8k (Mistral NeMo Minitron 8B).

On HumanEval, LLaMA 3.1 leads at 89% vs Mistral NeMo Minitron 8B at 36.22%. On ARC-Challenge, LLaMA 3.1 leads at 96.9% vs Mistral NeMo Minitron 8B at 64.42%. On MMLU, LLaMA 3.1 leads at 87.3% vs Mistral NeMo Minitron 8B at 69.51%. On GSM8K, LLaMA 3.1 leads at 96.8% vs Mistral NeMo Minitron 8B at 58.45%.