Mixtral 8x7B Instruct v2

NEW HOT

Mistral AI · Mistral

Open SourceAPI

An updated Mixture-of-Experts model, offering high-quality responses and efficient inference for demanding applications.

TextCodeReasoning

Scores 80% on MMLU, competitive across 4 standard benchmarks.

Data aggregated and editorially reviewed by TrendMing.

Benchmark Performance

MMLU80%
HumanEval79%
MATH56%
GPQA41%

Scores sourced from published evaluations. Higher = better.

Quick Specs

Context Window64K tokens
Parameters47B (effective)
Input pricingN/A (self-hosted)
Output pricingN/A (self-hosted)
Released2026-01-10