| GPT-4o | 88.1% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| Claude 3.5 Sonnet | 87.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| DeepSeek-V3 | 87.2% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| DeepSeek-V3 | 87.1% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| Gemini 1.5 Pro | 86.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 3.5 Sonnet | 86.5% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-4 | 86.4% | Self-reported | GPT-4 Technical Report via Epoch AI | — |
| Llama 3.3 70B | 86.3% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| Gemini 1.5 Pro | 85.9% | Third party | Gemini 1.5 Report via Epoch AI | — |
| Qwen2.5-72B-Instruct | 85.3% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| Phi-4 | 84.8% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| Claude 3 Opus | 84.6% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3.1 405B | 84.5% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3.1 405B | 84.4% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| GPT-4o | 84.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-4o | 84.2% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Qwen2.5-72B-Instruct | 83.4% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-4 | 82.4% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Qwen2-72B-Instruct | 82.4% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Amazon Nova Pro | 82.0% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-4o mini | 81.8% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| GPT-4 Turbo | 81.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3.2 90B Vision | 80.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3.1 70B | 80.1% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Mistral Large 2.0 | 80.0% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Qwen2.5-14B-Instruct | 79.9% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| Gemini 2.0 Flash | 79.7% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-4 Turbo | 79.6% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3 70B | 79.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Yi-Large | 79.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3.3 70B | 79.1% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 2 | 78.5% | Self-reported | Model Card and Evaluations for Claude Models via Epoch AI | — |
| DeepSeek-V2 | 78.4% | Self-reported | DeepSeek-V3 Technical Report via Epoch AI | — |
| Phi-3-medium | 78.0% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Gemini 1.5 Flash | 77.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Phi-3-medium | 77.9% | Self-reported | Phi-4 Technical Report via Epoch AI | — |
| Phi-3-medium | 77.5% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Amazon Nova Lite | 77.0% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 1 | 77.0% | Self-reported | Model Card and Evaluations for Claude Models via Epoch AI | — |
| GPT-4o mini | 76.7% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 3 Sonnet | 75.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Gemma 2 27B | 75.7% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 3.5 Haiku | 74.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Gemini 1.5 Flash | 73.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 3 Haiku | 73.8% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude 2.1 | 73.5% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Yi-34B-Chat | 73.5% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Claude Instant 1 | 73.4% | Self-reported | Model Card and Evaluations for Claude Models via Epoch AI | — |
| Claude Instant 1 | 73.4% | Self-reported | Releasing Claude Instant 1.2 via Epoch AI | — |
| Claude Instant 1.2 | 73.2% | Self-reported | Releasing Claude Instant 1.2 via Epoch AI | — |
| Qwen2.5-7B-Instruct | 72.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Inflection-1 | 72.7% | Self-reported | Inflection-1 via Epoch AI | — |
| Gemma 2 9B | 72.1% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-3.5 Turbo | 71.4% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Amazon Nova Micro | 70.8% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Falcon-180B Chat | 70.6% | Self-reported | The Falcon Series of Open Language Models via Epoch AI | — |
| Gemini 1.0 Pro | 70.0% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| text-davinci-002 | 70.0% | Self-reported | GPT-4 Technical Report via Epoch AI | — |
| Command R+ (08-2024) | 69.4% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| GPT-3.5 Turbo | 68.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Claude Instant 1.2 | 68.8% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3 8B | 68.8% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Mistral Large 1.0 | 68.8% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Mistral Small 1.0 | 68.7% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Stable Beluga 2 (Llama 2 70B fine-tune) | 68.6% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Yi-9B | 68.4% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Yi-34B-Chat | 67.6% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| GPT-3.5 Turbo | 67.3% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3 8B | 66.5% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Gemma 7B | 66.1% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Command R (08-2024) | 65.2% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Qwen-14B-Chat | 65.0% | Third party | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Gemma 7B | 64.3% | Self-reported | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Gemma 7B | 64.3% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Qwen-14B-Chat | 64.0% | Third party | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Gemma 7B | 63.6% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Mistral 7B v0.1 | 62.5% | Third party | Mixtral of Experts via Epoch AI | — |
| Mistral 7B v0.2 | 62.5% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Mistral 7B v0.1 | 61.7% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Yi-6B-Chat | 61.0% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Mistral 7B v0.1 | 60.1% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Mistral 7B v0.3 | 59.9% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Falcon 2 11B | 58.4% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| Phi-2 | 58.4% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Yi-6B-Chat | 58.2% | Self-reported | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| InternLM-Chat-20B | 57.4% | Third party | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Falcon-40B Instruct | 56.9% | Self-reported | Falcon2-11B Technical ReportFalcon2-11B Technical Report via Epoch AI | — |
| Mistral 7B v0.1 | 56.6% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 3.2 11B Vision | 56.5% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Phi-2 | 56.3% | Self-reported | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone via Epoch AI | — |
| Llama 3.1 8B | 56.1% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| InternLM-Chat-20B | 55.6% | Third party | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Llama 2 13B | 55.6% | Third party | Mixtral of Experts via Epoch AI | — |
| Falcon-40B Instruct | 55.4% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Falcon-40B Instruct | 55.4% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Llama 2 13B | 55.4% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Baichuan2-13B-Chat | 55.1% | Third party | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| Llama 2 13B | 55.1% | Third party | Baichuan 2: Open Large-scale Language Models via Epoch AI | — |
| Llama 2 13B | 54.8% | Self-reported | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 13B | 54.8% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Llama 2 13B | 54.8% | Self-reported | Nemotron-4 15B Technical Report via Epoch AI | — |
| Baichuan2-13B-Chat | 50.1% | Third party | Yi: Open Foundation Models by 01.AI via Epoch AI | — |
| ChatGLM2-6B | 47.9% | Self-reported | Qwen Technical Report via Epoch AI | — |
| MPT-30B (Instruct, Chat) | 47.9% | Self-reported | Qwen Technical Report via Epoch AI | — |
| MPT-30B (Instruct, Chat) | 46.9% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 7B | 45.8% | Third party | Stanford CRFM Leaderboard via Epoch AI | — |
| Llama 2 7B | 45.7% | Third party | Baichuan 2: Open Large-scale Language Models via Epoch AI | — |
| Llama 2 7B | 45.3% | Self-reported | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Llama 2 7B | 45.3% | Third party | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Llama 2 7B | 45.3% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Llama 2 7B | 44.4% | Third party | Mixtral of Experts via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 43.9% | Third party | Scaling Language Models: Methods, Analysis & Insights from Training Gopher via Epoch AI | — |
| text-davinci-001 (InstructGPT) | 43.9% | Third party | LLaMA: Open and Efficient Foundation Language Models via Epoch AI | — |
| Baichuan-7B (base) | 42.3% | Self-reported | Baichuan 2: Open Large-scale Language Models via Epoch AI | — |
| Gemma 2B | 42.3% | Self-reported | Gemma: Open Models Based on Gemini Research and Technology via Epoch AI | — |
| Phi-1.5 | 37.6% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Falcon-7B Instruct | 35.0% | Self-reported | Falcon2-11B Technical Report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 30.8% | Self-reported | Qwen Technical Report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 27.9% | Third party | Baichuan 2: Open Large-scale Language Models via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 27.4% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 26.9% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| Falcon-7B Instruct | 26.9% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 26.8% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 26.8% | Self-reported | Textbooks Are All You Need II: phi-1.5 technical report via Epoch AI | — |
| MPT-7B (Instruct, Chat, StoryWriter) | 26.7% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Dolly 2.0 (3B, 7B, 12B) | 26.2% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 26.2% | Third party | Llama 2: Open Foundation and Fine-Tuned Chat Models via Epoch AI | — |
| Falcon-7B Instruct | 26.2% | Self-reported | Qwen Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 26.0% | Third party | Baichuan 2: Open Large-scale Language Models via Epoch AI | — |
| Dolly 2.0 (3B, 7B, 12B) | 25.4% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| OPT-13B | 25.1% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| OPT-13B | 24.4% | Self-reported | XGen-7B Technical Report via Epoch AI | — |
| Falcon-7B Instruct | 23.9% | Self-reported | XGen-7B Technical Report via Epoch AI | — |