Luminia-13B-v3: benchmark results

Provider: Other. Access: Open.

Unified ELO 1380 ± 20, rank #2289 of 2928 rated models, from 12 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard - MuSR39.83Score40.8
Open LLM Leaderboard v1 - MMLU53.6Accuracy (%) (5-shot)33
Open LLM Leaderboard v1 - WinoGrande72.61Accuracy (%) (5-shot)29.5
Open LLM Leaderboard v1 - HellaSwag76.08Normalized accuracy (%) (10-shot)27.9
Open LLM Leaderboard v1 - ARC Challenge52.47Normalized accuracy (%) (25-shot)27.2
Open LLM Leaderboard - BBH41.12Score25.9
Open LLM Leaderboard - GPQA27.01Score24.8
Open LLM Leaderboard v1 - TruthfulQA MC243.74MC2 (%) (0-shot)24.7
Open LLM Leaderboard - MMLU-Pro22.15Score23
Open LLM Leaderboard v1 - GSM8K4.25Accuracy (%) (5-shot)22.6
Open LLM Leaderboard - IFEval25.23Score21.1
Open LLM Leaderboard - MATH Level 51.81Score12.9

Interactive version: theaggregate.ai/model?slug=luminia-13b-v3 · How It Works · Data refreshed daily, snapshot 2026-09-23.