Mistral: Mixtral 8x22B Instruct

MISTRALAI Developer Architecture Profile

Intelligence (ELO)1220Chatbot Arena Verified

Max Context65,536Tokens

API Cost / 1M$8.00Blended Prompt + Completion

Model Capabilities

Classification
Conversational

Mistral's official instruct fine-tuned version of [Mixtral 8x22B](/models/mistralai/mixtral-8x22b). It uses 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. Its strengths include: - strong math, coding, and reasoning - large context length (64k) - fluency in English, French, Italian, German, and Spanish See benchmarks on the launch announcement [here](https://mistral.ai/news/mixtral-8x22b/). #moe

Granular Pricing Matrix

Input Tokens (Prompt)$2.00 / 1M

Output Tokens (Completion)$6.00 / 1M

Pricing data via OpenRouter. Sync: 3/16/2026

Evaluate Competitors

VS Engine MatchupMistral: Mixtral 8x22B Instruct vs Writer: Palmyra X5 VS Engine MatchupMistral: Mixtral 8x22B Instruct vs OpenAI: GPT Audio VS Engine MatchupMistral: Mixtral 8x22B Instruct vs Qwen: Qwen3 Max VS Engine MatchupMistral: Mixtral 8x22B Instruct vs AI21: Jamba Large 1.7 VS Engine MatchupMistral: Mixtral 8x22B Instruct vs Cohere: Command A VS Engine MatchupMistral: Mixtral 8x22B Instruct vs Perplexity: Sonar Deep Research