Kaiju-A-57B: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1435 ± 40, rank #4422 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU72.66Accuracy (%) (5-shot)95.4
Open LLM Leaderboard v1 - WinoGrande78.77Accuracy (%) (5-shot)65.4
Open LLM Leaderboard v1 - TruthfulQA MC252.29MC2 (%) (0-shot)55.2
Open LLM Leaderboard v1 - GSM8K38.36Accuracy (%) (5-shot)53.3
Open LLM Leaderboard v1 - HellaSwag80.95Normalized accuracy (%) (10-shot)43
Open LLM Leaderboard v1 - ARC Challenge58.79Normalized accuracy (%) (25-shot)40.9

Interactive version: theaggregate.ai/model?slug=kaiju-a-57b · How It Works · Data refreshed daily, snapshot 2026-11-08.