mpt-7B-chat — benchmark results

Provider: Databricks. Released 2023-05-22. Access: Open.

Unified ELO 1210 ± 9, rank #1674 of 1776 rated models, from 75 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LLM Trustworthy - Fairness100Trust Score (%)94
MMLU-by-task - College Mathematics36Accuracy (%)80.3
MMLU-by-task - Virology42.77Accuracy (%)60.2
MMLU-by-task - Electrical Engineering45.52Accuracy (%)56
MMLU-by-task - Machine Learning33.04Accuracy (%)53.1
MMLU-by-task - Abstract Algebra30Accuracy (%)52.1
LLM Trustworthy - Adversarial46.2Trust Score (%)52
LLM Trustworthy - Out-of-Distribution64.26Trust Score (%)52
MMLU-by-task - Security Studies48.57Accuracy (%)48.8
LLM Trustworthy - Privacy78.93Trust Score (%)48
MMLU-by-task - College Chemistry32Accuracy (%)46.7
MMLU-by-task - High School Computer Science41Accuracy (%)46.2

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