Midnight-Rose-70B-v2.0.3: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1459 ± 35, rank #3359 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU69.64Accuracy (%) (5-shot)92.8
Open LLM Leaderboard v1 - ARC Challenge70.65Normalized accuracy (%) (25-shot)86.6
Open LLM Leaderboard v1 - HellaSwag87.5Normalized accuracy (%) (10-shot)86.5
Open LLM Leaderboard v1 - TruthfulQA MC265.27MC2 (%) (0-shot)83.5
Open LLM Leaderboard v1 - WinoGrande81.22Accuracy (%) (5-shot)77.8
Open LLM Leaderboard v1 - GSM8K28.35Accuracy (%) (5-shot)45.9

Interactive version: theaggregate.ai/model?slug=midnight-rose-70b-v2-0-3 · How It Works · Data refreshed daily, snapshot 2026-11-08.