Infinirc-Llama3-8B-2G-Release-v1.0 Benchmark Results

Infinirc-Llama3-8B-2G-Release-v1.0 — benchmark results

Provider: Other. Released 2024-06-26. Access: Open.

Unified ELO 1261 ± 73, rank #1597 of 1776 rated models, from 14 benchmark results.

Benchmark	Score	Metric	Percentile
Open LLM Leaderboard - MuSR	16.75	Score	86.7
Open LLM Leaderboard - GPQA	6.6	Score	55.6
Open LLM Leaderboard - BBH	20.83	Score	31.1
Open LLM Leaderboard - MMLU-Pro	12.89	Score	22.7
Open Chinese LLM - CMMLU	44.65	Accuracy (%)	12.2
Open Chinese LLM - WinoGrande	59.59	Accuracy (%)	12.2
Open LLM Leaderboard - IFEval	20.24	Score	12.2
Open LLM Leaderboard - MATH Level 5	1.66	Score	12.1
Open Chinese LLM - TruthfulQA MC	47.39	Accuracy (%)	10.2
Open Chinese LLM - C-Eval Semantic	49.05	Accuracy (%)	9.5
Open Chinese LLM - HellaSwag	50.51	Accuracy (%)	8.9
Open Chinese LLM - ARC Challenge	39.16	Accuracy (%)	6.2