Metric: Accuracy (higher is better)
| # | Model↕ | Accuracy▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | FLAN 137B (zero-shot) | 78.4 | No | Finetuned Language Models Are Zero-Shot Learners | 2021-09-03 | Code |
| 2 | FLAN 137B (few-shot, k=16) | 78.2 | No | Finetuned Language Models Are Zero-Shot Learners | 2021-09-03 | Code |
| 3 | LLaMA 65B (zero-shot) | 60.2 | No | LLaMA: Open and Efficient Foundation Language Mo... | 2023-02-27 | Code |
| 4 | LLaMA 33B (zero-shot) | 58.6 | No | LLaMA: Open and Efficient Foundation Language Mo... | 2023-02-27 | Code |
| 5 | GPT-3 175B (zero-shot) | 57.6 | No | Language Models are Few-Shot Learners | 2020-05-28 | Code |
| 6 | LLaMA 7B (zero-shot) | 57.2 | No | LLaMA: Open and Efficient Foundation Language Mo... | 2023-02-27 | Code |
| 7 | LLaMA 13B (zero-shot) | 56.4 | No | LLaMA: Open and Efficient Foundation Language Mo... | 2023-02-27 | Code |
| 8 | PaLM 540B (zero-shot) | 53.4 | No | PaLM: Scaling Language Modeling with Pathways | 2022-04-05 | Code |
| 9 | PaLM 62B (zero-shot) | 50.4 | No | PaLM: Scaling Language Modeling with Pathways | 2022-04-05 | Code |