ruGPT-3-small: benchmark results

Provider: Other. Access: Open.

Unified ELO 1169 ± 39, rank #1481 of 1537 rated models, from 31 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MERA Original Paper - ruMMLU26.3Accuracy (%; 5-shot, log-likelihood)61.1
MERA Original Paper - ruWorldTree25.7Accuracy (%; 5-shot, log-likelihood)61.1
MERA Original Paper - ruModAr0.1Exact Match (%; greedy generation)58.3
MERA Original Paper - ruMultiAr0.9Exact Match (%; 5-shot, greedy generation)50
MERA Original Paper - ruOpenBookQA25.8Accuracy (%; 5-shot, log-likelihood)50
MERA - LCS10.4Accuracy (%)29.2
MERA Original Paper - SimpleAr0Exact Match (%; 5-shot, greedy generation)22.2
MERA - USE8.14Grade, normalized (%)21.3
MERA - ruHateSpeech54.34Accuracy (%)13.2
MERA Original Paper - MathLogicQA24.4Accuracy (%; 5-shot, log-likelihood)11.1
MERA - RWSD46.15Accuracy (%)10.5
MERA Original Paper19.1Total Score (%; mean of 17 problem-solving and exam tasks, M8.3

Interactive version: theaggregate.ai/model?slug=rugpt-3-small · How It Works · Data refreshed daily, snapshot 2026-09-25.