OPT-175B: benchmark results

Provider: Meta. Released 2022-05-03. Access: Open.

Unified ELO 1310 ± 34, rank #2421 of 2656 rated models, from 36 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
HELM Classic - XSUM15.49ROUGE-2 (%)92.7
HELM Classic - OpenbookQA58.6Exact Match (%)88.7
HELM Classic - HellaSwag79.1Exact Match (%)77.4
HELM Classic - LSAT22.03Exact Match (%)77.2
HELM Classic - Synthetic Reasoning Natural24.79F1 (%)76.5
HELM Classic - MS MARCO Regular28.78RR@10 (%)72.4
HELM Classic - BLiMP83.05Exact Match (%)71
HELM Classic - BoolQ79.3Exact Match (%)69.7
HELM Classic - IMDB94.73Exact Match (%)68.9
HELM Classic - CNN/DailyMail14.59ROUGE-2 (%)65.9
HELM Classic - bAbI50.66Exact Match (%)65.2
HELM60.95Mean win rate (self-reported)61.5

Interactive version: theaggregate.ai/model?slug=opt-175b · How It Works · Data refreshed daily, snapshot 2026-09-19.