Mistral Nemo — benchmark results

Mistral's 12B Apache-2.0 model built with NVIDIA, with a 128K context, designed as a drop-in upgrade for Mistral 7B. Provider: Mistral. Released 2024-07-18. Access: Open.

Unified ELO 1470 ± 10, rank #907 of 1776 rated models, from 341 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AGC-Bench - puntuguese1.32Dataset z-score97.6
AGC-Bench - hypobench0.43Dataset z-score96.3
LIBRA - ruQasper24.47Dataset Total Score (%)92.9
LIBRA - ruSciPassageCount12.83Dataset Total Score (%)92.9
AGC-Bench - creatset1.3Dataset z-score86.6
AGC-Bench - c3_crosstalk0.89Dataset z-score79.3
LIBRA - MatreshkaNames32.22Dataset Total Score (%)78.6
LIBRA - ruTREC58.4Dataset Total Score (%)78.6
AGC-Bench - humor_transfer0.88Dataset z-score77.2
AGC-Bench - pun_eval0.66Dataset z-score76.8
AGC-Bench - ocw_connections0.88Dataset z-score75.6
LIBRA - ruBABILongQA316Dataset Total Score (%)71.4

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