Mixtral 8x7B Instruct

Bedrock · Mistral · Standard

Approved GA

Mistral AI's sparse mixture-of-experts model with 8 experts of 7B parameters each, delivering strong performance at fast inference speeds.

Context window

32K

Max output

N/A

Latency

Low

Cost / 1K tokens

Variable

Available on

Platform	Model ID	Cost / 1K	Approval
Bedrock Primary	mixtral-8x7b-instruct	Variable	Approved

Text Code

Chat Code generation Reasoning Multilingual tasks