vicuna-class-shishya-all-hal-13B-ep3: benchmark results

Provider: LMSYS. Access: Open.

Aggregate Rating 1341 ± 35, rank #7937 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - GSM8K25.85Accuracy (%) (5-shot)44.3
Open LLM Leaderboard v1 - HellaSwag80.28Normalized accuracy (%) (10-shot)40.5
Open LLM Leaderboard v1 - MMLU56.4Accuracy (%) (5-shot)39
Open LLM Leaderboard v1 - WinoGrande73.16Accuracy (%) (5-shot)31.2
Open LLM Leaderboard v1 - ARC Challenge48.63Normalized accuracy (%) (25-shot)23.4
Open LLM Leaderboard v1 - TruthfulQA MC242.75MC2 (%) (0-shot)21.5

Interactive version: theaggregate.ai/model?slug=vicuna-class-shishya-all-hal-13b-ep3 · How It Works · Data refreshed daily, snapshot 2026-11-08.