Open LLM Leaderboard (China Mirror)
Open LLM Leaderboard tracks model performance on ARC, HellaSwag, MMLU, TruthfulQA, Winogrande, and GSM8K benchmarks.
Top Model
h2ogpt-gm-oasst1-en-1024-20b
Top Score
-
Model Count
100
Data version
-
Data source: HuggingFace
Leaderboard snapshot month:
Ranking Table
| Model | Type | Parameters (B) | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K | Architecture |
|---|---|---|---|---|---|---|---|---|---|---|
| h2ogpt-gm-oasst1-en-1024-20b | Fine Tuned Models | 200 | 42.58 | 48.04 | 72.76 | 25.96 | 39.92 | 66.30 | 2.50 |
Data is for reference only. Official sources are authoritative. Click model names to view DataLearner model profiles.