Mixtral 8x7B

Training compute

7.7×10²³ FLOP

Parameters

46.7B

Published

Dec 11, 2023

Mixtral 8x7B is an AI model developed by Mistral AI (France), first published in December 2023. It works in the language domain, on tasks such as language modeling/generation, question answering, quantitative reasoning and translation.

Training it took an estimated 7.7×10²³ FLOP of compute (estimation method: operation counting). The model has 46,700,000,000 parameters.

Access: Open weights (unrestricted). Its weights are openly available. Epoch AI rates the confidence of this record as speculative.

Full record

Organization: Mistral AI
Country of organization: France
Domain: Language
Task: Language modeling/generation, Question answering, Quantitative reasoning, Translation
Training compute: 7.7×10²³ FLOP
Compute estimation method: Operation counting
Parameters: 46,700,000,000
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Epoch confidence: Speculative

Benchmark results

01

Epoch Capabilities Index

ECI Score

118.1