Pallas-0.4 [bfloat16] [rev: 2cef301d3afa]: benchmark results

Provider: Mihaiii. Access: Open.

Aggregate Rating 1468 ± 37, rank #2898 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU74.93Accuracy (%) (5-shot)96.4
Open LLM Leaderboard v1 - WinoGrande80.43Accuracy (%) (5-shot)73.8
Open LLM Leaderboard v1 - GSM8K60.88Accuracy (%) (5-shot)73.1
Open LLM Leaderboard v1 - TruthfulQA MC257.26MC2 (%) (0-shot)69.7
Open LLM Leaderboard v1 - ARC Challenge63.65Normalized accuracy (%) (25-shot)63.7
Open LLM Leaderboard v1 - HellaSwag83.3Normalized accuracy (%) (10-shot)57.7

Interactive version: theaggregate.ai/model?slug=pallas-0-4-bfloat16-rev-2cef301d3afa · How It Works · Data refreshed daily, snapshot 2026-11-08.