MBX-7B-v3: benchmark results

Provider: Other. Access: Open.

Aggregate Rating 1475 ± 36, rank #2745 of 10916 rated models, from 7 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - ARC Challenge74.15Normalized accuracy (%) (25-shot)99
Open LLM Leaderboard v1 - WinoGrande85.56Accuracy (%) (5-shot)98.6
Open LLM Leaderboard v1 - HellaSwag88.91Normalized accuracy (%) (10-shot)95
Open LLM Leaderboard v1 - TruthfulQA MC271.87MC2 (%) (0-shot)92.5
Open LLM Leaderboard v1 - GSM8K70.28Accuracy (%) (5-shot)92.3
YALL Nous Leaderboard57.14Average83.5
Open LLM Leaderboard v1 - MMLU65.06Accuracy (%) (5-shot)81.1

Interactive version: theaggregate.ai/model?slug=mbx-7b-v3 · How It Works · Data refreshed daily, snapshot 2026-11-08.