Skip to content
Metatext Hub — one governed catalog for your company’s AI toolsGet early access
Metatext
Sign Up

Mixtral 8x22B Instruct

Prime InferenceOpen weights

Mixtral 8x22B Instruct by Prime Inference costs $2.20 per 1M input tokens and $6.60 per 1M output tokens, with a 66K-token context window.

Pricing

Input (per 1M tokens)
$2.20
Output (per 1M tokens)
$6.60
Cached input (per 1M)
—

Specifications

Provider
Prime Inference
Context window
66K tokens
Parameters
—
Released
Apr 2024
Open weights
Yes
Frontier model
No

Compare Mixtral 8x22B Instruct with…

Mixtral 8x22B Instruct vs Claude Opus 4.6$25.00/1MMixtral 8x22B Instruct vs Claude Opus 4.7$25.00/1MMixtral 8x22B Instruct vs Claude Opus 4.8$25.00/1MMixtral 8x22B Instruct vs Claude Opus 5$25.00/1MMixtral 8x22B Instruct vs Claude Fable 5.1$50.00/1MMixtral 8x22B Instruct vs Claude Fable 5$50.00/1M

FAQ

Pricing is per 1M tokens (USD); confirm with the provider before production use.