provider,model,release_date,epoch_eci,gpqa_epoch,lmarena_elo,mmlu,source_urls OpenAI,GPT-4 (Mar 2023),2023-03-14,125.74,0.3573,1288.0,0.864,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude 1.3,2023-04-18,,,,0.77,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) Google,PaLM 2 (Bison),2023-05-10,,,1139.3,0.6921,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,GPT-3.5 Turbo (Jun 2023),2023-06-13,113.22,,,0.6888,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4 (Jun 2023),2023-06-13,123.09,0.3065,1276.3,0.8244,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Inflection,Inflection-1,2023-06-22,,,,0.727,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) Anthropic,Claude 2,2023-07-11,120.19,0.3466,,0.785,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Meta,Llama 2-13B,2023-07-18,105.58,,1141.5,0.5542,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 2-34B,2023-07-18,104.93,,,0.626,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Meta,Llama 2-70B,2023-07-18,113.67,0.2633,1170.7,0.6955,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 2-7B,2023-07-18,98.61,,1108.0,0.4581,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Instant,2023-08-09,120.21,,,0.6879,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,Phi-1.5,2023-09-11,91.06,,,0.376,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen-14B,2023-09-24,112.88,,1139.5,0.663,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral 7B v0.1,2023-09-27,112.05,,1110.5,0.5658,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral 7B v0.3,2023-09-27,108.88,0.1518,,0.5991,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen-7B,2023-09-28,106.59,,,0.45,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) 01.AI,Yi 6B,2023-11-02,104.5,,,0.6401,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" 01.AI,Yi-34B,2023-11-02,117.33,0.1474,,0.7623,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek Coder 1.3B,2023-11-02,62.59,,,0.258,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) DeepSeek,DeepSeek Coder 33B,2023-11-02,95.94,,,0.394,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) DeepSeek,DeepSeek Coder 6.7B,2023-11-02,89.13,,,0.364,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) OpenAI,GPT-3.5 Turbo (Nov 2023),2023-11-06,118.53,0.2803,1204.4,0.714,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4 Turbo (Nov 2023),2023-11-06,126.43,0.4236,1313.0,0.7955,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude 2.1,2023-11-21,119.32,0.3295,,0.7348,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek LLM 67B,2023-11-29,,0.2462,1185.1,0.725,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Mistral 7B v0.2,2023-12-11,,,1149.5,0.625,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Mistral Medium,2023-12-11,,,1222.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Mixtral 8x7B,2023-12-11,118.44,0.3059,1197.3,0.7173,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,Phi-2,2023-12-12,107.73,,,0.5842,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 1.0 Pro,2023-12-13,117.14,0.3396,1236.9,0.7,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-3.5 Turbo (Jan 2024),2024-01-25,115.68,0.2718,1225.7,0.6727,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen1.5-14B,2024-02-04,,,1191.2,0.6864,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen1.5-32B,2024-02-04,,0.3074,1204.1,0.7443,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen1.5-72B,2024-02-04,,0.2882,1233.6,0.7739,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen1.5-7B,2024-02-04,,,1144.0,0.6258,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 1.5 Pro (Feb 2024),2024-02-15,,,,0.8095,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot)" Google,Gemma 2B,2024-02-21,93.79,,1093.7,0.423,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 7B,2024-02-21,111.82,,1137.8,0.6612,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral Large,2024-02-26,122.12,0.3876,1242.7,0.6878,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral Small (Feb 2024),2024-02-26,,,,0.6873,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models" NVIDIA,Nemotron-4 15B,2024-02-26,107.49,,,0.587,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Anthropic,Claude 3 Opus,2024-02-29,126.9,0.4716,1322.3,0.8457,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude 3 Sonnet,2024-02-29,120.84,0.4059,1281.4,0.7592,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude 3 Haiku,2024-03-07,118.85,0.363,1261.8,0.738,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Cohere,Command R (Mar 2024),2024-03-12,,,1227.0,0.652,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Cohere,Command R+ (Apr 2024),2024-04-04,,,1261.9,0.6935,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,GPT-4 Turbo (Apr 2024),2024-04-09,127.27,0.4659,1324.5,0.8127,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,WizardLM-2 8x22B,2024-04-15,,0.4343,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Mistral,Mixtral 8x22B,2024-04-17,122.04,0.3406,1229.7,0.7782,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3-70B,2024-04-18,123.01,0.4056,1276.6,0.7934,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3-8B,2024-04-18,116.39,0.2607,1223.7,0.6676,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,phi-3-medium 14B,2024-04-23,121.19,0.2759,1198.0,0.7745,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,phi-3-mini 3.8B,2024-04-23,117.28,,1128.2,0.688,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,phi-3-small 7.4B,2024-04-23,121.82,,1171.2,0.7568,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen1.5-110B,2024-04-25,,,1234.3,0.7678,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" DeepSeek,"DeepSeek-V2 (MoE-236B, May 2024)",2024-05-07,124.8,,,0.784,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://web.archive.org/web/20250101082016/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) 01.AI,Yi-1.5-34B,2024-05-13,,0.3198,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) 01.AI,Yi-Large,2024-05-13,,,,0.7932,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models" OpenAI,GPT-4o (May 2024),2024-05-13,128.99,0.489,1346.1,0.8417,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 1.5 Flash (May 2024),2024-05-14,122.68,0.4037,1286.6,0.779,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 1.5 Pro (May 2024),2024-05-14,126.94,0.4586,1324.1,0.8271,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-4 (0520),2024-05-20,,,1273.5,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen2-72B,2024-06-07,125.28,0.4078,1261.9,0.8235,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" NVIDIA,Nemotron-4 340B,2024-06-14,,,1277.3,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" DeepSeek,DeepSeek-Coder-V2 236B,2024-06-17,,,1265.4,,"https://web.archive.org/web/20250101082016/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Anthropic,Claude 3.5 Sonnet,2024-06-20,130.0,0.5404,1343.3,0.8654,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 2 27B,2024-06-24,122.12,0.3649,1289.5,0.757,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 2 9B,2024-06-24,119.81,0.2746,1267.0,0.7213,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral NeMo,2024-07-18,118.65,0.2989,,0.6533,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4o mini,2024-07-18,126.46,0.3772,1317.6,0.7672,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3.1-405B,2024-07-23,128.77,0.5092,1335.5,0.8453,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3.1-70B,2024-07-23,125.93,0.4419,1293.4,0.801,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3.1-8B,2024-07-23,116.47,0.2696,1211.2,0.5606,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral Large 2 (Jul 2024),2024-07-24,127.57,0.4902,1314.4,0.8,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4o (Aug 2024),2024-08-06,128.83,0.4921,1335.6,0.8433,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok Beta,2024-08-13,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models xAI,Grok-2 (Aug 2024),2024-08-13,,,1335.7,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" xAI,Grok-2 mini,2024-08-13,,,1308.4,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 1.5 Flash (Sep 2024),2024-08-27,129.33,0.4732,1309.3,0.7386,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Cohere,Command R (Aug 2024),2024-08-30,,,1250.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Cohere,Command R+,2024-08-30,119.2,,1276.4,0.694,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V2.5 (Sep 2024),2024-09-06,,,1307.2,,"https://web.archive.org/web/20250101082016/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,o1-mini,2024-09-12,135.7,0.6237,1336.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,o1-preview,2024-09-12,134.57,0.5032,1388.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen2.5-32B,2024-09-17,,0.4609,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Mistral,Mistral Small v24.09,2024-09-17,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Mistral,Pixtral 12B,2024-09-17,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Alibaba (Qwen),Qwen2.5-Coder (1.5B),2024-09-18,102.71,,,0.536,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen2.5-Coder (7B),2024-09-18,113.01,,,0.68,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen2.5-Coder-32B,2024-09-18,119.45,,,0.791,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen-Max (Sep 2024),2024-09-19,,,1318.1,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen2.5-14B,2024-09-19,,,,0.799,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) Alibaba (Qwen),Qwen2.5-72B,2024-09-19,128.97,0.4915,1302.9,0.8337,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen2.5-7B,2024-09-19,118.54,0.3548,,0.7287,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 1.5 Pro (Sept 2024),2024-09-24,131.72,0.5723,1351.3,0.8688,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3.2 11B,2024-09-24,,,,0.5651,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24)" Meta,Llama 3.2 1B,2024-09-24,,0.2393,1111.2,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Meta,Llama 3.2 3B,2024-09-24,,,1166.9,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Meta,Llama 3.2 90B,2024-09-24,125.52,0.4104,,0.8027,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 1.5 Flash 8B,2024-10-03,,0.3295,1258.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" NVIDIA,Llama 3.1 Nemotron 70B Instruct,2024-10-15,,,1298.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Ministral 3B,2024-10-16,118.04,0.2525,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Mistral,Ministral 8B,2024-10-16,,0.2715,1237.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Anthropic,Claude 3.5 Haiku,2024-10-22,127.07,0.3813,1324.8,0.7432,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude 3.5 Sonnet (October 2024),2024-10-22,133.43,0.553,1374.3,0.8727,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Cohere,Aya Expanse 32B,2024-10-24,,,1267.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen-Turbo,2024-11-01,,0.4179,,,https://web.archive.org/web/20250402114911/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Mistral,Mistral Large 2 (Nov 2024),2024-11-15,128.57,0.5133,1305.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Pixtral Large,2024-11-18,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models OpenAI,GPT-4o (Nov 2024),2024-11-20,128.69,0.4789,,0.881,https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen2.5-Coder-32B-Instruct,2024-11-21,,,1270.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen2.5-Plus (Nov 2024),2024-11-27,,,1314.6,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),QwQ-32B-Preview,2024-11-28,,,1154.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Amazon,Amazon Nova Lite,2024-12-03,,,1260.4,0.7702,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Amazon,Amazon Nova Micro,2024-12-03,,,1240.7,0.7076,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Amazon,Amazon Nova Pro,2024-12-03,,,1290.1,0.82,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 2.0 Flash (Dec 2024),2024-12-06,134.58,,,0.7974,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 3.3 70B,2024-12-06,127.33,0.4744,1318.1,0.7906,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V2.5 (Dec 2024),2024-12-10,,,1323.4,,"https://web.archive.org/web/20250101082016/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Microsoft,Phi-4,2024-12-12,130.48,0.5606,1256.3,0.848,"https://epoch.ai/data/benchmark_data.zip (mmlu_external.csv) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok-2 (Dec 2024),2024-12-12,130.47,0.5379,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) OpenAI,o1,2024-12-17,141.72,0.7677,1402.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V3,2024-12-26,132.3,0.5653,1358.4,0.872,"https://crfm.stanford.edu/helm/mmlu/v1.13.0/ (HELM MMLU all subjects, 5-shot) | https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Codestral,2025-01-13,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models MiniMax,MiniMax-Text-01,2025-01-14,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models DeepSeek,DeepSeek-R1,2025-01-20,138.82,0.7172,1398.1,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-R1-Distill-Llama-70B,2025-01-20,,0.5574,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) DeepSeek,DeepSeek-R1-Distill-Llama-8B,2025-01-20,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models DeepSeek,DeepSeek-R1-Distill-Qwen-1.5B,2025-01-20,,0.3359,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) DeepSeek,DeepSeek-R1-Distill-Qwen-14B,2025-01-20,,0.447,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) DeepSeek,DeepSeek-R1-Distill-Qwen-32B,2025-01-20,,0.6414,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Google,Gemini 2.0 Flash Thinking (Jan 2025),2025-01-21,135.24,0.5707,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen Plus (Jan 2025),2025-01-25,,0.4811,1345.9,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen2.5-Max,2025-01-25,132.48,0.5612,1373.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral Small 3,2025-01-25,,0.4729,1274.4,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,o3-mini,2025-01-31,140.2,0.7702,1363.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.0 Flash (Feb 2025),2025-02-05,134.62,0.6414,1360.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.0 Flash-Lite,2025-02-05,,,1353.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 2.0 Pro,2025-02-05,134.95,0.6566,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Mistral,Mistral Saba,2025-02-17,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Anthropic,Claude 3.7 Sonnet,2025-02-24,141.03,0.7973,1388.4,,"https://web.archive.org/web/20250901150210/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Microsoft,Phi-4 Multimodal,2025-02-26,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Microsoft,Phi-4 Mini,2025-02-27,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) OpenAI,GPT-4.5,2025-02-27,136.65,0.6869,1444.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),QwQ-32B,2025-03-05,137.5,0.6534,1335.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 3 27B,2025-03-11,129.97,0.4773,1365.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 3 12B,2025-03-12,123.54,0.3946,1341.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 3 1B,2025-03-12,,0.1995,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Google,Gemma 3 4B,2025-03-12,115.88,0.2323,1303.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Cohere,Cohere Command A,2025-03-13,,,1353.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Mistral Small 3.1,2025-03-17,127.53,0.4747,1302.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V3 (Mar 2025),2025-03-24,135.85,0.6761,1395.2,,"https://web.archive.org/web/20250901150210/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Pro (Mar 2025),2025-03-25,143.96,0.8384,,,https://web.archive.org/web/20250402114911/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) OpenAI,GPT-4o (Mar 2025),2025-03-27,,,1442.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Meta,Llama 4 Maverick,2025-04-05,132.22,0.6698,1327.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Llama 4 Scout,2025-04-05,129.63,0.5183,1321.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" NVIDIA,Llama 3.1 Nemotron Ultra 253B,2025-04-07,,,1347.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),QWQ-Plus,2025-04-08,,0.654,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) xAI,Grok 3,2025-04-09,138.21,0.7576,1411.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok-3 mini,2025-04-09,140.2,0.7626,1363.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4.1,2025-04-14,136.65,0.6692,1414.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4.1 mini,2025-04-14,134.89,0.6585,1382.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-4.1 nano,2025-04-14,129.54,0.4893,1322.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,o3,2025-04-16,146.73,0.8182,1431.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,o4-mini,2025-04-16,145.46,0.7961,1391.1,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Flash (Apr 2025),2025-04-17,139.84,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen3-14B,2025-04-28,138.15,0.6376,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen3-235B-A22B,2025-04-28,139.15,0.7071,1402.1,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3-30B-A3B,2025-04-28,136.11,0.6168,1326.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3-32B,2025-04-28,138.42,0.6572,1346.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3-8B,2025-04-28,136.06,0.5676,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen3-0.6B,2025-04-29,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Alibaba (Qwen),Qwen3-1.7B,2025-04-29,,0.3801,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Alibaba (Qwen),Qwen3-4B,2025-04-29,,0.5234,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) Amazon,Nova Premier,2025-04-30,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Google,Gemini 2.5 Pro (May 2025),2025-05-06,142.15,0.6667,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Mistral,Mistral Medium 3,2025-05-07,134.02,0.5953,1387.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Flash (May 2025),2025-05-20,141.43,,,,https://web.archive.org/web/20250603043751/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Mistral,Devstral Small,2025-05-21,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Anthropic,Claude Opus 4,2025-05-22,142.52,0.7626,1426.3,,"https://web.archive.org/web/20250901150210/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Sonnet 4,2025-05-22,141.56,0.7919,1401.9,,"https://web.archive.org/web/20250901150210/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-R1 (May 2025),2025-05-28,141.13,0.7633,1421.4,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Pro (Jun 2025),2025-06-05,145.13,0.8529,1445.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Magistral Medium 1.1,2025-06-10,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Mistral,Magistral Small 1.0,2025-06-10,133.07,0.5606,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) OpenAI,o3-pro,2025-06-10,147.33,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) MiniMax,MiniMax-M1-80k,2025-06-13,,,1363.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 2.5 Flash (Jun 2025),2025-06-17,140.59,,1409.2,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Flash-Lite (Jun 2025),2025-06-17,133.82,,1374.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" MiniMax,MiniMax-M1-40k,2025-06-17,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Mistral,Mistral Small 3.2,2025-06-20,,0.4905,1356.1,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemma 3n E4B,2025-06-26,,,1317.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" xAI,Grok 4,2025-07-09,146.24,0.87,1410.9,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Devstral Medium,2025-07-10,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Moonshot (Kimi),Kimi K2 (Jul 2025),2025-07-12,139.97,,1417.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-4.5-Air,2025-07-20,,,1373.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3-235B-A22B-Instruct (Jul 2025),2025-07-25,138.78,,1422.2,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3-235B-A22B-Thinking (Jul 2025),2025-07-25,143.76,0.8005,1399.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" NVIDIA,Llama Nemotron Super 49B v1.5,2025-07-25,,,1343.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3-30B-A3B-Instruct (Jul 2025),2025-07-29,137.27,0.5562,1382.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3-30B-A3B-Thinking (Jul 2025),2025-07-30,139.55,0.7008,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen3-Coder-480B-A35B,2025-07-31,,,1387.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Zhipu (Z.ai),GLM-4.5,2025-08-03,,,1410.9,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Anthropic,Claude Opus 4.1,2025-08-05,144.07,0.7727,1449.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,gpt-oss-120b,2025-08-05,139.79,0.7576,1351.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,gpt-oss-20b,2025-08-05,137.7,0.608,1317.4,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5,2025-08-07,150.0,0.8617,1434.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5 mini,2025-08-07,145.34,0.75,1390.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5 nano,2025-08-07,139.21,0.6944,1337.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-4.5V,2025-08-11,,,1352.2,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Mistral Medium 3.1,2025-08-12,,,1407.8,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemma 3 270M,2025-08-14,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) NVIDIA,Nemotron Nano 9B v2,2025-08-18,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) ByteDance,Seed-OSS-36B,2025-08-20,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models DeepSeek,DeepSeek-V3.1,2025-08-21,139.79,,1417.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok Code Fast 1,2025-08-28,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Moonshot (Kimi),Kimi K2 (Sep 2025),2025-09-05,,,1418.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3-Next-80B-A3B,2025-09-11,,,1398.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,GPT-5-Codex,2025-09-15,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models Mistral,Magistral Medium 1.2,2025-09-18,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Mistral,Magistral Small 1.2,2025-09-18,,0.476,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) xAI,Grok 4 Fast,2025-09-19,144.09,,1418.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V3.1-Terminus,2025-09-22,,,1416.2,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3-VL-235B-A22B,2025-09-23,,,1413.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3-Max,2025-09-24,142.33,0.726,1423.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Flash (Sep 2025),2025-09-25,142.9,,1403.0,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 2.5 Flash-Lite (Sep 2025),2025-09-25,,,1379.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Anthropic,Claude Sonnet 4.5,2025-09-29,146.74,0.8232,1456.4,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V3.2-Exp,2025-09-29,144.91,,1425.2,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-4.6,2025-09-30,,,1424.3,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,GPT-5 Pro,2025-10-07,150.59,,,,https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Anthropic,Claude Haiku 4.5,2025-10-15,142.15,0.7121,1414.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" MiniMax,MiniMax-M2,2025-10-27,,,1343.6,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Moonshot (Kimi),Kimi Linear 48B A3B,2025-10-30,,,,,https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Moonshot (Kimi),Kimi K2 Thinking,2025-11-06,145.82,0.8422,1430.1,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.1,2025-11-13,149.76,0.8763,1455.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok 4.1,2025-11-17,,,1465.2,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 3 Pro,2025-11-18,153.07,0.9261,1485.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok 4.1 Fast,2025-11-19,,,1430.5,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Anthropic,Claude Opus 4.5,2025-11-24,150.11,0.8605,1473.7,,"https://web.archive.org/web/20251201192238/https://artificialanalysis.ai/leaderboards/models | https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V3.2,2025-12-01,146.09,0.8342,1424.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Amazon,Nova 2.0 Lite,2025-12-02,,,1335.0,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Amazon,Nova 2.0 Omni,2025-12-02,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Amazon,Nova 2.0 Pro Preview,2025-12-02,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Mistral,Ministral 3 14B,2025-12-02,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Mistral,Ministral 3 3B,2025-12-02,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Mistral,Ministral 3 8B,2025-12-02,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Mistral,Mistral Large 3,2025-12-02,,,1414.3,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" OpenAI,GPT-5.2,2025-12-11,153.66,0.914,1437.4,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.2 Pro,2025-12-11,156.33,,,,https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) NVIDIA,Nemotron 3 Nano 30B A3B,2025-12-15,,,1313.7,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Xiaomi,MiMo-V2-Flash,2025-12-16,,,1391.6,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Google,Gemini 3 Flash,2025-12-17,151.88,0.8939,1472.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-4.7,2025-12-22,143.59,0.8333,1441.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" MiniMax,MiniMax-M2.1,2025-12-23,,,1384.6,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Moonshot (Kimi),Kimi K2.5,2026-01-27,148.07,0.876,1449.9,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" StepFun,Step 3.5 Flash,2026-02-02,,,1393.0,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Anthropic,Claude Opus 4.6,2026-02-05,155.66,0.9053,1504.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.3 Codex,2026-02-05,157.09,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Zhipu (Z.ai),GLM-5,2026-02-11,145.8,0.8782,1457.4,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" MiniMax,MiniMax-M2.5,2026-02-12,146.59,,1390.7,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.5 397B-A17B,2026-02-13,146.8,0.8636,1441.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen 3.5 Plus (hosted 397B-A17B),2026-02-16,146.17,0.8485,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Anthropic,Claude Sonnet 4.6,2026-02-17,152.79,0.8737,1472.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok 4.20,2026-02-17,152.11,0.8933,1474.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.1 Pro,2026-02-19,155.34,0.9444,1487.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.5-0.8B,2026-02-24,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Alibaba (Qwen),Qwen3.5-122B-A10B,2026-02-24,,,1416.6,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3.5-27B,2026-02-24,,,1408.7,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3.5-2B,2026-02-24,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Alibaba (Qwen),Qwen3.5-35B-A3B,2026-02-24,142.49,0.8346,1394.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.5-4B,2026-02-24,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Alibaba (Qwen),Qwen3.5-9B,2026-02-24,139.32,0.7898,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen 3.5 Flash (hosted 35B-A3B),2026-02-25,143.52,0.8232,1396.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.1 Flash-Lite,2026-03-03,144.26,0.8182,1433.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.4,2026-03-05,157.31,0.933,1475.4,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.4 Pro,2026-03-05,159.57,0.946,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) NVIDIA,Nemotron 3 Super,2026-03-11,,,1360.1,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Mistral,Mistral Small 4,2026-03-16,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) OpenAI,GPT-5.4 Mini,2026-03-17,148.7,0.8687,1447.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.4 Nano,2026-03-17,146.03,0.7847,1401.4,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" MiniMax,MiniMax-M2.7,2026-03-18,145.78,,1415.0,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Xiaomi,MiMo-V2-Pro,2026-03-18,,,1448.0,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen 3.6 Plus,2026-03-31,147.3,0.8838,1443.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 4 26B A4B,2026-04-02,141.88,0.7323,1437.5,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 4 31B IT,2026-04-02,142.99,0.7576,1452.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-5.1,2026-04-07,149.82,0.899,1465.4,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Spark,2026-04-08,152.03,0.898,1489.2,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen 3.6 35B-A3B,2026-04-14,143.73,0.8485,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Anthropic,Claude Opus 4.7,2026-04-16,156.79,0.9015,1502.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" xAI,Grok 4.3 Beta,2026-04-17,148.81,0.8883,1441.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen 3.6 Max (Preview),2026-04-20,149.41,0.8737,1460.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Moonshot (Kimi),Kimi K2.6,2026-04-20,150.94,0.9078,1460.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.6 27B,2026-04-22,146.23,0.8586,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) OpenAI,GPT-5.5,2026-04-23,159.68,0.94,1481.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.5 Pro,2026-04-23,162.91,0.9392,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Xiaomi,MiMo-V2.5,2026-04-23,,,1434.0,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Xiaomi,MiMo-V2.5-Pro,2026-04-23,,,1467.3,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" DeepSeek,DeepSeek-V4-Flash,2026-04-24,146.11,,1438.7,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek-V4-Pro,2026-04-24,149.13,0.9091,1457.9,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen 3.6 Flash,2026-04-27,142.89,0.8333,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Mistral,Mistral Medium 3.5,2026-04-28,141.36,,1426.5,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.5 Instant,2026-05-05,142.43,0.8251,1473.2,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.7-Max,2026-05-19,154.04,0.9091,1474.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.5 Flash,2026-05-19,154.88,0.928,1479.5,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Cohere,Command A+,2026-05-20,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Anthropic,Claude Opus 4.8,2026-05-28,159.01,0.9104,1481.5,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" MiniMax,MiniMax-M3,2026-06-01,146.87,0.9091,1441.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.7-Plus,2026-06-02,147.0,0.8788,1455.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemma 4 12B,2026-06-03,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) NVIDIA,Nemotron 3 Ultra,2026-06-04,145.97,0.8535,1426.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Fable 5,2026-06-09,163.42,0.8586,1505.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Moonshot (Kimi),Kimi K2.7 Code,2026-06-12,150.11,0.8788,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Zhipu (Z.ai),GLM-5.2,2026-06-16,151.94,0.9186,1475.9,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Sonnet 5,2026-06-30,156.88,0.9053,1464.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Tencent,Hy3,2026-07-06,,,1453.8,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" xAI,Grok 4.5,2026-07-08,154.07,0.9343,1465.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Spark 1.1,2026-07-09,154.05,,1490.2,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.6 Luna,2026-07-09,156.5,0.916,1453.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.6 Sol,2026-07-09,162.79,0.935,1485.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-5.6 Terra,2026-07-09,160.09,0.9331,1467.2,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Moonshot (Kimi),Kimi K3,2026-07-16,157.86,0.9312,1488.1,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.5 Flash-Lite,2026-07-21,144.91,0.8333,1455.2,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.6 Flash,2026-07-21,154.53,0.9413,1483.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Opus 5,2026-07-24,163.94,0.9388,1490.9,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.7 Flash,2026-07-27,144.05,0.8232,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) DeepSeek,DeepSeek V4 Flash 0731,2026-07-31,154.76,0.9104,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Alibaba (Qwen),Qwen 3.8 Max,2026-08-02,157.4,0.9268,1482.9,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Spark 1.2,2026-08-05,155.05,,1494.6,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Glimmer,2026-08-10,,,1424.7,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Alibaba (Qwen),Qwen3.8 2.4T A95B,2026-08-12,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) xAI,Grok 4.6,2026-08-12,156.57,0.94,1456.4,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek V4 Pro 0813,2026-08-13,155.64,0.9167,1464.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.7 Flash,2026-08-13,157.97,0.9482,1489.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen 3.8 27B,2026-08-14,151.69,,1437.8,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-5.3,2026-08-14,155.99,0.9091,1480.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Zhipu (Z.ai),GLM-5.3-Flash,2026-08-20,153.08,0.9015,1476.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Alibaba (Qwen),Qwen3.8 Max (0902),2026-09-01,155.08,0.923,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) Anthropic,Claude Fable 5.1,2026-09-01,166.49,,1503.2,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 3.8 Flash,2026-09-02,157.44,0.9539,1496.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Spark 1.3,2026-09-02,157.21,,1495.6,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-6 Astra,2026-09-03,168.81,0.9577,1477.3,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" DeepSeek,DeepSeek V4.1 Flash,2026-09-09,155.93,0.8984,1476.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" StepFun,Step 5 Preview,2026-09-18,,,1447.7,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Xiaomi,MiMo-V2.6 Flash,2026-09-21,,,1450.9,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Xiaomi,MiMo-V2.6 Pro,2026-09-21,,,1479.5,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" xAI,Grok 4.7,2026-09-21,154.37,0.9274,1444.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Opus 5.5,2026-09-22,168.83,0.9059,1508.7,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-6 Luna,2026-09-22,156.75,0.9047,1442.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-6 Sol,2026-09-22,163.73,0.9432,1459.0,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Sonnet 5.5,2026-09-28,166.18,0.9558,1477.5,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" OpenAI,GPT-6.1 Sol,2026-09-29,168.37,0.9539,1483.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Google,Gemini 4 Argon,2026-09-30,166.45,0.964,1524.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Mistral,Mistral Large 4,2026-10-06,150.03,0.8807,1433.8,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Anthropic,Claude Haiku 5.5,2026-10-07,159.32,0.8965,1424.6,,"https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Wonder,2026-10-12,161.85,,1512.7,,"https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv)" Meta,Muse Spark Open,2026-10-15,155.11,0.8857,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) | https://epoch.ai/data/benchmark_data.zip (epoch_capabilities_index/eci_scores.csv) DeepSeek,DeepSeek V4.1 Flash 1017,2026-10-17,,0.9318,,,https://epoch.ai/data/benchmark_data.zip (gpqa_diamond.csv) | https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) MiniMax,MiniMax-M3.1-Flash,2026-10-19,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) Moonshot (Kimi),Kimi K3.1 Pro,2026-10-23,,,,,https://artificialanalysis.ai/leaderboards/models (retrieved 2026-10-24) ByteDance,Dola Seed 2.0 Pro,,,,1456.3,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)" Zhipu (Z.ai),GLM-4-Plus,,,,1319.4,,"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset (text_style_control, overall)"