ZMIME
Mistral AI · Released Apr 17, 2024

Mixtral 8x22B

Mistral model for multilingual chat, reasoning, and tool-assisted workflows.

  • Open weights
  • Tool calling
Capability
122.0ECI · #139 of 148
Input
$2.00per 1M tokens
Output
$6.00per 1M tokens
Context
64K64K max output
Pricing

What it costs

Input
$2.00per million tokens
Output
$6.00per million tokens
Blended (3:1)
$3.00pricier than 74% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$7.00
Team assistant25M input + 5M output tokens$80.00
Production app250M input + 50M output tokens$800.00

Official Mistral API price, as listed on models.dev.

Capability

Independent benchmarks

122.0Epoch Capabilities Index
#139 of 148 scored models
88Median 146167

The shaded band is Epoch AI’s confidence range (115.1–124.4); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions34.1%

Scores from Epoch AI, run independently of Mistral AI.

Specifications

The details

Lab
Mistral AI
Released
Apr 17, 2024
Knowledge cutoff
Apr 2024
Context window
64,000 tokens
Max output
64,000 tokens
Inputs
Text
Output
Text
Reasoning
No
Tool calling
Yes
Structured output
No
Weights
Open weights
API model ID
open-mixtral-8x22bon Mistral
Availability
1 API providerlisted on models.dev

Mistral model documentation

Same lab

More from Mistral AI

Questions

About Mixtral 8x22B

How much does Mixtral 8x22B cost?

Mixtral 8x22B costs $2.00 input / $6.00 output per million tokens (official Mistral API price). At a 3:1 input-to-output mix that is $3.00 per million tokens, more expensive than 74% of the 360 priced models we track.

What is the context window of Mixtral 8x22B?

Mixtral 8x22B accepts up to 64,000 tokens per request and can write up to 64,000 tokens in one response.

How good is Mixtral 8x22B?

Epoch AI gives Mixtral 8x22B a Capabilities Index score of 122.0 (likely range 115.1–124.4), ranking it #139 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 34.1%.

Is Mixtral 8x22B open source?

Yes. Mistral AI publishes the weights, so it can be downloaded and self-hosted.

What inputs does Mixtral 8x22B support?

Mixtral 8x22B accepts text and replies in text. It is not a dedicated reasoning model, supports tool calling.

When was Mixtral 8x22B released?

Mistral AI released Mixtral 8x22B on Apr 17, 2024. Its training data runs to Apr 2024.

What are the best alternatives to Mixtral 8x22B?

The closest current models from other labs on capability, price and release date are GPT-4o (OpenAI, ECI 129.0, $2.50 / $10.00), Nova Pro (Amazon, ECI 123.8, $0.80 / $3.20), Qwen2.5 72B Instruct (Alibaba (Qwen), ECI 129.0, $1.40 / $5.60) and Llama-3.1-70B-Instruct (Meta, ECI 125.9, $0.72 / $0.72).