| StableLM-Alpha 7B | 51.5% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Mistral 7B v0.1 | 54.2% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Phi-2 | 54.7% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Gemma 7B | 55.6% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| GPT-2 XL (1.5B) | 58.3% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| GPT-2 XL (1.5B) | 58.3% | Source not stated | Epoch AI via Epoch AI | — |
| OPT-1.3B | 61.0% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Dolly 2.0 (3B, 7B, 12B) | 61.8% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| OPT-13B | 64.7% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Llama 3 8B | 65.0% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Gemma 2B | 65.4% | Self-reported | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Falcon-7B Instruct | 66.2% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Falcon-7B Instruct | 66.3% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Falcon-7B Instruct | 67.2% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 67.2% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 67.2% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 68.0% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 68.3% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 68.6% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| GPT-3.5 Turbo | 68.8% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Llama 2 7B | 69.1% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Llama 2 7B | 69.2% | Self-reported | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 7B | 69.2% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 70.2% | Third party | LLaMA: Open and Efficient Foundation Language Models via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 70.2% | Third party | Language Models are Few-Shot Learners via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 70.2% | Third party | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 70.2% | Third party | Language Models are Few-Shot Learners via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 70.2% | Third party | Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model via Epoch AI | — |
| MPT-30B (Instruct, Chat) | 71.0% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Gemma 7B | 72.3% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Llama 2 13B | 72.8% | Self-reported | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 13B | 72.8% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Llama 2 13B | 72.8% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Yi-9B | 73.0% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 73.2% | Third party | Language Models are Few-Shot Learners via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 73.2% | Third party | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 73.2% | Third party | Language Models are Few-Shot Learners via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 73.2% | Third party | Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model via Epoch AI | — |
| Phi-1.5 | 73.4% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Claude 3 Haiku | 74.2% | Self-reported | The Claude 3 Model Family: Opus, Sonnet, Haiku via Epoch AI | — |
| Mistral 7B v0.1 | 74.2% | Third party | Mixtral of Experts via Epoch AI | — |
| Mistral 7B v0.1 | 74.2% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Claude 3 Sonnet | 75.1% | Self-reported | The Claude 3 Model Family: Opus, Sonnet, Haiku via Epoch AI | — |
| Mistral 7B v0.1 | 75.3% | Third party | Training Compute-Optimal Large Language Models via Epoch AI | — |
| Mistral 7B v0.1 | 75.3% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Falcon-40B Instruct | 76.4% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon-40B Instruct | 76.4% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon-40B Instruct | 76.9% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 77.7% | Third party | Language Models are Few-Shot Learners via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 77.7% | Third party | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 77.7% | Third party | Language Models are Few-Shot Learners via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 77.7% | Third party | Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model via Epoch AI | — |
| Falcon 2 11B | 78.3% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon 2 11B | 78.3% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Gemma 7B | 79.0% | Self-reported | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Phi-3-medium | 81.5% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| GPT-3.5 Turbo | 81.6% | Third party | The Falcon Series of Open Language Models via Epoch AI | — |
| text-davinci-002 | 81.6% | Self-reported | GPT-4 technical report via Epoch AI | — |
| Llama 3.1 405B | 82.2% | Self-reported | The Llama 3 Herd of Models via Epoch AI | — |
| DeepSeek-V3 | 85.2% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| DeepSeek-V2 | 86.3% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| Falcon-180B Chat | 87.1% | Self-reported | The Falcon Series of Open Language Models via Epoch AI | — |
| GPT-4 | 87.5% | Self-reported | GPT-4 technical report via Epoch AI | — |
| GPT-4 32K | 87.5% | Third party | The Falcon Series of Open Language Models via Epoch AI | — |
| Claude 3 Opus | 88.5% | Self-reported | The Claude 3 Model Family: Opus, Sonnet, Haiku via Epoch AI | — |
| Llama 3.1 405B | 89.2% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |