Skip to content
Tokenly.
Demo fixture — not an offerDemo fixture

Mixtral 8x7B Instruct

mistralai/Mixtral-8x7B-Instruct-v0.1

A sparse mixture-of-experts model with 46B total parameters and 32K-token context, known for efficient throughput on hosted infrastructure.

Mixtral 8x7B Instruct
Model IDmistralai/Mixtral-8x7B-Instruct-v0.1
DeveloperMistral AI
Serving providerPending confirmation
Capabilitieschat
Modalitiestext
Context window33K tokens
Pricing
Input
Not yet published
Output
Not yet published
LimitationsEarlier generation than Mixtral 8x22B; instruction quality trails current flagships.
SourceModel card ↗
Last verified18 Nov 2026