Skywork-o1-Open-Llama-3.1-8B Benchmark Results

Skywork-o1-Open-Llama-3.1-8B — benchmark results

Provider: Skywork. Released 2024-11-26. Access: Open.

Unified ELO 1276 ± 99, rank #1580 of 1776 rated models, from 12 benchmark results.

Benchmark	Score	Metric	Percentile
Open LLM Leaderboard - MATH Level 5	52.11	Score	97.8
Open LLM Leaderboard - BBH	23.02	Score	34.8
Open LLM Leaderboard - IFEval	35.18	Score	33.7
Open LLM Leaderboard - MMLU-Pro	11.45	Score	21.8
Open LLM Leaderboard - GPQA	1.23	Score	11.8
Open Persian LLM - ARC Challenge	41.61	Accuracy (%)	10
Open Persian LLM - ARC Easy	47.12	Accuracy (%)	10
Open Persian LLM - AUT Multiple Choice Persian	40.43	Accuracy (%)	10
Open Persian LLM Leaderboard	34.15	Average Score (%)	10
Open LLM Leaderboard - MuSR	1.52	Score	6
Open Persian LLM - MMLU Pro	14.55	Accuracy (%)	3.3
Open Persian LLM - Part Multiple Choice	27.02	Accuracy (%)	3.3