Beast-Soul-new: benchmark results

Provider: Other. Access: Open.

Unified ELO 1504 ± 21, rank #1166 of 2928 rated models, from 12 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - WinoGrande85.24Accuracy (%) (5-shot)97.8
Open LLM Leaderboard v1 - ARC Challenge73.12Normalized accuracy (%) (25-shot)95.8
Open LLM Leaderboard v1 - HellaSwag88.35Normalized accuracy (%) (10-shot)91.2
Open LLM Leaderboard v1 - GSM8K69.75Accuracy (%) (5-shot)90.3
Open LLM Leaderboard v1 - TruthfulQA MC267.38MC2 (%) (0-shot)85.9
Open LLM Leaderboard - MuSR44.86Score82.6
Open LLM Leaderboard v1 - MMLU64.74Accuracy (%) (5-shot)75.8
Open LLM Leaderboard - BBH52.27Score60.5
Open LLM Leaderboard - IFEval50.3Score59.4
Open LLM Leaderboard - MMLU-Pro31.08Score40.4
Open LLM Leaderboard - MATH Level 57.4Score38.9
Open LLM Leaderboard - GPQA28.27Score37.6

Interactive version: theaggregate.ai/model?slug=beast-soul-new · How It Works · Data refreshed daily, snapshot 2026-09-23.