← Back to Directory

Mistral: Mixtral 8x7B Instruct

MISTRALAI Developer Architecture Profile

OpenRouter ↗HuggingFace ↗
Intelligence (ELO)1114Chatbot Arena Verified
Max Context32,768Tokens
API Cost / 1M$1.08Blended Prompt + Completion

Model Capabilities

  • Classification
  • Conversational
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 parameters. Instruct model fine-tuned by Mistral. #moe

Granular Pricing Matrix

Input Tokens (Prompt)$0.54 / 1M
Output Tokens (Completion)$0.54 / 1M

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

Evaluate Competitors

VS Engine MatchupMistral: Mixtral 8x7B Instruct vs Z.ai: GLM 5 TurboVS Engine MatchupMistral: Mixtral 8x7B Instruct vs Inception: Mercury 2VS Engine MatchupMistral: Mixtral 8x7B Instruct vs Qwen: Qwen3.5-27BVS Engine MatchupMistral: Mixtral 8x7B Instruct vs Qwen: Qwen3.5-122B-A10BVS Engine MatchupMistral: Mixtral 8x7B Instruct vs AionLabs: Aion-2.0VS Engine MatchupMistral: Mixtral 8x7B Instruct vs Qwen: Qwen3.5 Plus 2026-02-15