Skip to main content
Back to Explore

Mixtral 8x7B Instruct

Mistral·Released 2023-12-11
Open Source33K ctxMoE

About

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...

Pricing

Input

$0.45

per 1M tokens

Output

$0.70

per 1M tokens

Blended

$0.51

per 1M tokens

Cheaper than 51% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.51

Monthly

$15.36

vs. Similar Models

Apertus 70B InstructQ:0.0
$1.34+163%
Jamba 1.6 MiniQ:+0.2
$0.25-51%
Qwen3 1.7B (Reasoning)Q:+0.2
$0.40-22%
Command-R (Mar '24)Q:-0.3
$0.75+46%

Performance

Context Window

33K

tokens

Larger than 7% of models

Max Output

16K

tokens

50% of context

Context Window Comparison

Mistral: Saba
33KSame
Qwen2.5 Coder 7B Instruct
33KSame
LFM2 8B A1B
33KSame

Benchmarks

MMLU-Pro
38.7%
GPQA Diamond
29.2%
HLE
4.5%
LiveCodeBench
6.6%
SciCode
2.8%
TerminalBench HardNot evaluated
MATH-500
29.9%
AIME
0.0%
AIME 2025Not evaluated
IFBenchNot evaluated
Long Context RecallNot evaluated
Tau2Not evaluated
Market AverageTop Score

Open Source

Quick Compare

Similar Models

Compare all 7 models