Mistral 7B — benchmark results

Provider: Mistral. Released 2023-09-27. Access: Open.

Unified ELO 1348 ± 18, rank #1422 of 1776 rated models, from 50 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LIBRA - ru2WikiMultihopQA43.21Dataset Total Score (%)85.7
LIBRA - ruSciPassageCount8.17Dataset Total Score (%)85.7
The Wrong Kind of Right26.8MAR Disability (self-reported)83.3
LIBRA - ruGSM10011Dataset Total Score (%)71.4
LIBRA - ruQasper5.38Dataset Total Score (%)71.4
LIBRA - ruTREC42.46Dataset Total Score (%)71.4
T-Eval56Overall Score (%)60
LIBRA - MatreshkaNames16.33Dataset Total Score (%)57.1
MHGraphBench42.1Avg$_{All^\ast}$ (self-reported)50
RedCode61.7Score (%)50
DiBiMT48.31Avg Accuracy (%)47.4
LIBRA - LibrusecMHQA33.59Dataset Total Score (%)42.9

Interactive version: theaggregate.ai/model?slug=mistral-7b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.