YamshadowExperiment28-7B Benchmark Results

YamshadowExperiment28-7B — benchmark results

mlabonne's automerger SLERP of YamShadow-7B and yam-peleg's Experiment28-7B, briefly the top 7B on the Open LLM Leaderboard amid overfitting caveats. Provider: Other. Released 2024-03-18. Access: Open.

Unified ELO 1409 ± 14, rank #1180 of 1776 rated models, from 14 benchmark results.

Strongest benchmark results

Benchmark	Score	Metric	Percentile
Open Chinese LLM - TruthfulQA MC	63.65	Accuracy (%)	94.2
Open LLM Leaderboard - MuSR	12.69	Score	69.5
Open Chinese LLM - ARC Challenge	53.92	Accuracy (%)	60.5
Open Chinese LLM - HellaSwag	60.95	Accuracy (%)	60.1
Open LLM Leaderboard - BBH	31.98	Score	59.4
Open Chinese LLM - GSM8K	46.17	Accuracy (%)	52.7
Open Chinese LLM Leaderboard	56.6	Average Score (%)	49.3
Open Chinese LLM - WinoGrande	62.75	Accuracy (%)	47.8
Open LLM Leaderboard - GPQA	4.92	Score	41.7
Open LLM Leaderboard - IFEval	40.7	Score	40.8
Open LLM Leaderboard - MMLU-Pro	22.89	Score	38.5
Open LLM Leaderboard - MATH Level 5	6.12	Score	33.5

Interactive version: theaggregate.ai/model?slug=yamshadowexperiment28-7b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.