Cosmosis-3x34B: benchmark results

Provider: Weyaxi. Access: Open.

Aggregate Rating 1506 ± 34, rank #2257 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU77.25Accuracy (%) (5-shot)98.5
Open LLM Leaderboard v1 - GSM8K72.25Accuracy (%) (5-shot)97.5
Open LLM Leaderboard v1 - WinoGrande84.14Accuracy (%) (5-shot)92.1
Open LLM Leaderboard v1 - ARC Challenge69.71Normalized accuracy (%) (25-shot)84
Open LLM Leaderboard v1 - TruthfulQA MC263.82MC2 (%) (0-shot)81.1
Open LLM Leaderboard v1 - HellaSwag85.18Normalized accuracy (%) (10-shot)72.4

Interactive version: theaggregate.ai/model?slug=cosmosis-3x34b · How It Works · Data refreshed daily, snapshot 2026-11-08.