Llama 3 8B Synthia V3.5 Benchmark Results

Llama 3 8B Synthia V3.5 — benchmark results

Provider: Meta. Released 2024-05-17. Access: Open.

Unified ELO 1382 ± 11, rank #1305 of 1776 rated models, from 15 benchmark results.

Benchmark	Score	Metric	Percentile
Open Korean LLM Leaderboard	275.79	Average Score (%)	65.3
Open Chinese LLM - TruthfulQA MC	55.3	Accuracy (%)	62.6
Open LLM Leaderboard - IFEval	50.7	Score	60.4
Open Chinese LLM - WinoGrande	63.46	Accuracy (%)	57.3
Open LLM Leaderboard - MuSR	9.41	Score	45.7
Open LLM Leaderboard - BBH	27.54	Score	43.9
Open Chinese LLM - GSM8K	41.77	Accuracy (%)	41.8
Open Chinese LLM - CMMLU	51.92	Accuracy (%)	40.7
Open Chinese LLM - C-Eval Semantic	65.39	Accuracy (%)	39.5
Open LLM Leaderboard - MMLU-Pro	22.56	Score	37.3
Open LLM Leaderboard - MATH Level 5	6.57	Score	35.6
Open Chinese LLM Leaderboard	54.71	Average Score (%)	35