Llama 3 8B Chinese Chat: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1430 ± 27, rank #4672 of 10916 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 66.97 | Accuracy (%) (5-shot) | 88.6 |
| Open LLM Leaderboard v1 - GSM8K | 67.17 | Accuracy (%) (5-shot) | 83.1 |
| Open Chinese LLM - GSM8K | 51.78 | Accuracy (%) | 70.9 |
| Open Chinese LLM Leaderboard | 58.97 | Official Nonzero-Task Mean (%) | 64.5 |
| Open Chinese LLM - C-Eval Semantic | 72.07 | Accuracy (%) | 58.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 61.77 | Normalized accuracy (%) (25-shot) | 54.4 |
| Open Chinese LLM - CMMLU | 54.1 | Accuracy (%) | 53.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 51.41 | MC2 (%) (0-shot) | 51.8 |
| Open Chinese LLM - WinoGrande | 62.67 | Accuracy (%) | 46.9 |
| Open LLM Leaderboard v1 - WinoGrande | 75.22 | Accuracy (%) (5-shot) | 40.9 |
| Open LLM Leaderboard v1 - HellaSwag | 80.07 | Normalized accuracy (%) (10-shot) | 39.9 |
| Open Chinese LLM - ARC Challenge | 48.55 | Accuracy (%) | 34.9 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-chinese-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.