AI Hub
All papers
Architecture

Mistral 7B

Mistral AI·October 10, 2023

Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch

View on arXiv

TL;DR

Introduces Mistral 7B, which outperforms larger Llama 2 models using grouped-query and sliding-window attention for efficient inference.

Why it matters

Proved a well-designed 7B model could beat much larger ones, launching Mistral and reinvigorating small open models.

Related models

Related terms