| DeepSeek-V3 | 95.3% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| Llama 3.1 405B | 95.3% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| DeepSeek-V2 | 92.2% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| Phi-3-medium | 91.6% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| GPT-3.5 Turbo | 87.4% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Claude Instant 1.2 | 86.3% | Self-reported | Releasing Claude Instant 1.2 via Epoch AI | — |
| Stable Beluga 2 (Llama 2 70B fine-tune) | 86.1% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Claude Instant 1 | 85.7% | Self-reported | Releasing Claude Instant 1.2 via Epoch AI | — |
| text-davinci-002 | 85.2% | Self-reported | GPT-4 Technical Report via Epoch AI | — |
| Llama 3 8B | 82.8% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Mistral 7B v0.1 | 78.6% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Gemma 7B | 78.3% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Phi-2 | 75.9% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Falcon-180B Chat | 67.8% | Self-reported | The Falcon Series of Open Language Models via Epoch AI | — |
| Falcon-180B Chat | 63.7% | Self-reported | The Falcon Series of Open Language Models via Epoch AI | — |
| Falcon-40B Instruct | 61.9% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| ChatGLM2-6B | 61.0% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Llama 2 13B | 60.3% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Falcon-40B Instruct | 56.7% | Self-reported | The Falcon Series of Open Language Models via Epoch AI | — |
| Yi-9B | 55.6% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Mistral 7B v0.1 | 55.5% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Mistral 7B v0.1 | 54.9% | Third party | Mixtral of Experts via Epoch AI | — |
| Mistral 7B v0.1 | 54.9% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Falcon-40B Instruct | 54.7% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon-40B Instruct | 54.5% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Gemma 7B | 53.2% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Gemma 7B | 53.2% | Self-reported | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 53.2% | Third party | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 51.5% | Third party | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 51.4% | Third party | The Falcon Series of Open Language Models via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 51.4% | Third party | LLaMA: Open and Efficient Foundation Language Models via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 51.4% | Third party | GLaM: Efficient Scaling of Language Models with Mixture-of-Experts via Epoch AI | — |
| MPT-30B (Instruct, Chat) | 50.6% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 13B | 49.4% | Third party | The Falcon Series of Open Language Models via Epoch AI | — |
| Llama 2 13B | 49.4% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Llama 2 13B | 49.4% | Self-reported | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 13B | 49.4% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Llama 2 13B | 48.8% | Third party | Mixtral of Experts via Epoch AI | — |
| Falcon-7B Instruct | 47.9% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Llama 2 7B | 45.9% | Third party | The Falcon Series of Open Language Models via Epoch AI | — |
| Llama 2 7B | 45.9% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Llama 2 7B | 45.9% | Self-reported | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 7B | 45.9% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Falcon-7B Instruct | 44.5% | Self-reported | The Falcon Series of Open Language Models via Epoch AI | — |
| Phi-1.5 | 44.4% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Falcon-7B Instruct | 43.4% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 43.4% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Llama 2 7B | 43.4% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Llama 2 7B | 43.2% | Third party | Mixtral of Experts via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 42.6% | Self-reported | Qwen Technical Report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 42.6% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Falcon-7B Instruct | 42.4% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 42.4% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Gemma 2B | 42.1% | Self-reported | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 41.7% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 40.5% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Dolly 2.0 (3B, 7B, 12B) | 39.6% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 36.3% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| OPT-13B | 35.8% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| StableLM-Alpha 7B | 27.0% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| GPT-2 XL (1.5B) | 25.0% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| OPT-1.3B | 23.2% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |