WiNGPT2-Llama-3-8B-Base: benchmark results
Provider: WinGPT. Released 2024-06-01. Access: Open.
Aggregate Rating 1420 ± 49, rank #5178 of 10916 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Medical-LLM Leaderboard (BenchmarkList) | 72.1 | Medical average (%) | 88.9 |
| Open Chinese LLM - CMMLU | 56.71 | Accuracy (%) | 62.6 |
| Open Chinese LLM - C-Eval Semantic | 68.13 | Accuracy (%) | 45 |
| Open Chinese LLM - HellaSwag | 59.07 | Accuracy (%) | 42.3 |
| Open Chinese LLM Leaderboard | 52.68 | Official Nonzero-Task Mean (%) | 25.4 |
| Open Chinese LLM - GSM8K | 31.31 | Accuracy (%) | 24 |
| Open Chinese LLM - ARC Challenge | 46.33 | Accuracy (%) | 21.7 |
| Open Chinese LLM - WinoGrande | 60.46 | Accuracy (%) | 17.2 |
| Open Chinese LLM - TruthfulQA MC | 46.76 | Accuracy (%) | 7.4 |
Interactive version: theaggregate.ai/model?slug=wingpt2-llama-3-8b-base · How It Works · Data refreshed daily, snapshot 2026-11-08.