TokenDealHub
中文

Mistral: Mixtral 8x7B Instructmistralai

Bottom Line

Mistral: Mixtral 8x7B Instruct: $0.54/1M input, $0.54/1M output, 33K context, CPS grade B. Mixtral 8x7B Instruct is a pretrained generative Sparse Mixture of Experts, by Mistral AI, for chat and instruction use. Incorporates 8 experts (feed-forward networks) for a total of 47 billion...

API Pricing (per 1M tokens)

Input$0.54
Output$0.54
Cache Read—
Cache Write—
Blended (3:1)$0.54

Specs & Performance

Context Length33K
Max Output16K
Intelligence5
Coding—
Math—
Speed—
TTFT—
CPS GradeB
Input / Output Modalitiestext → text

More from mistralai

Mistral: Mistral Medium 3.1
$0.40 / $2.00
Mistral: Mistral Medium 3
$0.40 / $2.00
Mistral: Devstral Small 1.1
$0.10 / $0.30
Mistral: Saba
$0.20 / $0.60
Mistral: Mistral Medium 3.5
$1.50 / $7.50
Mistral Large
$2.00 / $6.00

Looking for cheaper alternatives?

Find the cheapest model by task, or compare two models head-to-head.

Cheapest by task →Compare two models →Compare all models →

FAQ

How much does the Mistral: Mixtral 8x7B Instruct API cost?

Mistral: Mixtral 8x7B Instruct costs $0.54/1M input tokens and $0.54/1M output tokens. Blended cost at a 3:1 ratio is about $0.54/1M tokens.

What is the context window of Mistral: Mixtral 8x7B Instruct?

Mistral: Mixtral 8x7B Instruct supports a 33K token context window with up to 16K output tokens.

Is Mistral: Mixtral 8x7B Instruct worth it?

Mistral: Mixtral 8x7B Instruct has a CPS grade of B (score 52), intelligence index 5, coding index —. Use the compare page to benchmark against rivals.