Kyllene-34B-v1.1: benchmark results

A MergeMonster roleplay merge released by TeeZee combining Bagel-DPO, Nous-Capybara, Nous-Hermes-2-Yi and SUS-Chat 34B releases with an extended RPG preset. Provider: TeeZee. Access: Open.

Aggregate Rating 1499 ± 38, rank #2372 of 10916 rated models, from 6 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Open LLM Leaderboard v1 - MMLU77.16Accuracy (%) (5-shot)98.3
Open LLM Leaderboard v1 - WinoGrande83.58Accuracy (%) (5-shot)89.7
Open LLM Leaderboard v1 - ARC Challenge68.77Normalized accuracy (%) (25-shot)81.3
Open LLM Leaderboard v1 - TruthfulQA MC262.95MC2 (%) (0-shot)79.9
Open LLM Leaderboard v1 - HellaSwag84.98Normalized accuracy (%) (10-shot)70.6
Open LLM Leaderboard v1 - GSM8K0Accuracy (%) (5-shot)4.5

Interactive version: theaggregate.ai/model?slug=kyllene-34b-v1-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.