Model,Display name,eci,eci_ci_low,eci_ci_high,date,Organization,Country (of organization),Model accessibility,Accessibility group,model_versions GPT-6.2 Astra,GPT-6.2 Astra,174.17,170.29,178.58,2026-11-17,OpenAI,United States of America,API access,Closed weights, Claude Fable 5.2,Claude Fable 5.2,173.5,170.09,177.36,2026-11-09,Anthropic,United States of America,API access,Closed weights, Gemini 4 Carbon,Gemini 4 Carbon,169.78,166.59,173.31,2026-11-10,Google DeepMind,United States of America,API access,Closed weights, GPT-6 Astra,GPT-6 Astra,169.2,166.33,172.48,2026-09-03,OpenAI,United States of America,API access,Closed weights, Claude Opus 5.5,Claude Opus 5.5,168.7,165.87,171.89,2026-09-22,Anthropic,United States of America,API access,Closed weights, GPT-6.1 Sol,GPT-6.1 Sol,168.63,165.7,171.91,2026-09-29,OpenAI,United States of America,API access,Closed weights, GPT-6.1 Terra,GPT-6.1 Terra,167.72,164.78,170.92,2026-10-26,OpenAI,United States of America,API access,Closed weights, Claude Fable 5.1,Claude Fable 5.1,167.09,164.43,170.11,2026-09-01,Anthropic,United States of America,API access,Closed weights, Claude Sonnet 5.5,Claude Sonnet 5.5,166.59,163.83,169.65,2026-09-28,Anthropic,United States of America,API access,Closed weights, Gemini 4 Argon,Gemini 4 Argon,166.57,163.94,169.51,2026-10-13,Google DeepMind,United States of America,API access,Closed weights, Claude Opus 5,Claude Opus 5,164.74,162.3,167.66,2026-07-24,Anthropic,United States of America,API access,Closed weights, Claude Fable 5,Claude Fable 5,164.21,161.78,166.86,2026-06-09,Anthropic,United States of America,API access,Closed weights, GPT-6 Sol,GPT-6 Sol,163.82,161.4,166.57,2026-09-22,OpenAI,United States of America,API access,Closed weights, GPT-5.6 Sol,GPT-5.6 Sol,163.05,160.76,165.59,2026-07-09,OpenAI,United States of America,API access,Closed weights, GPT-5.5 Pro,GPT-5.5 Pro,162.77,160.14,165.77,2026-04-23,OpenAI,United States of America,API access,Closed weights, Muse Wonder,Muse Wonder,161.74,159.36,164.32,2026-10-12,Meta AI,United States of America,API access,Closed weights, Kimi K3.1 Pro,Kimi K3.1 Pro,161.36,158.97,163.9,2026-10-23,Moonshot,China,Open weights (restricted use),Open weights, GPT-5.6 Terra,GPT-5.6 Terra,160.78,158.58,163.23,2026-07-09,OpenAI,United States of America,API access,Closed weights, GPT-5.4 Pro,GPT-5.4 Pro,160.32,157.87,163.03,2026-03-05,OpenAI,United States of America,API access,Closed weights, Claude Haiku 5.5,Claude Haiku 5.5,160.29,158.12,162.69,2026-10-07,Anthropic,United States of America,API access,Closed weights, GPT-5.5,GPT-5.5,160.19,158.2,162.43,2026-04-23,OpenAI,United States of America,API access,Closed weights, Claude Opus 4.8,Claude Opus 4.8,159.91,157.82,162.21,2026-05-28,Anthropic,United States of America,API access,Closed weights, DeepSeek V4.1 Flash 1017,DeepSeek V4.1 Flash 1017,159.4,157.16,161.93,2026-10-17,DeepSeek,China,Open weights (unrestricted),Open weights, Gemini 3.8 Flash,Gemini 3.8 Flash,158.83,156.86,160.97,2026-09-02,Google DeepMind,United States of America,API access,Closed weights, Kimi K3,Kimi K3,158.6,156.63,160.8,2026-07-16,Moonshot,China,Open weights (non-commercial),Open weights, Gemini 3.7 Flash,Gemini 3.7 Flash,158.38,156.41,160.53,2026-08-13,Google DeepMind,United States of America,API access,Closed weights, Muse Spark 1.3,Muse Spark 1.3,158.25,156.26,160.41,2026-09-02,Meta AI,United States of America,API access,Closed weights, GPT-5.4,GPT-5.4,157.87,155.96,159.97,2026-03-05,OpenAI,United States of America,API access,Closed weights, Grok 4.6,Grok 4.6,157.82,155.86,160.0,2026-08-12,xAI,United States of America,API access,Closed weights, Qwen 3.8 Max,Qwen 3.8 Max,157.59,155.47,159.94,2026-08-02,Alibaba,China,API access,Closed weights, Claude Opus 4.7,Claude Opus 4.7,157.5,155.63,159.46,2026-04-16,Anthropic,United States of America,API access,Closed weights, GPT-5.2 Pro,GPT-5.2 Pro,157.31,155.11,159.77,2025-12-11,OpenAI,United States of America,API access,Closed weights, GLM-5.3,GLM-5.3,157.1,155.04,159.33,2026-08-14,Z.ai (Zhipu AI),China,Open weights (restricted use),Open weights, DeepSeek V4.1 Flash,DeepSeek V4.1 Flash,157.0,155.08,159.03,2026-09-09,DeepSeek,China,Open weights (unrestricted),Open weights, Claude Sonnet 5,Claude Sonnet 5,156.9,154.91,159.08,2026-06-30,Anthropic,United States of America,API access,Closed weights, GPT-5.6 Luna,GPT-5.6 Luna,156.7,154.87,158.77,2026-07-09,OpenAI,United States of America,API access,Closed weights, Grok 4.7,Grok 4.7,156.43,154.47,158.52,2026-09-21,xAI,United States of America,API access,Closed weights, GPT-6 Luna,GPT-6 Luna,156.39,154.5,158.47,2026-09-22,OpenAI,United States of America,API access,Closed weights, Claude Opus 4.6,Claude Opus 4.6,156.35,154.59,158.29,2026-02-05,Anthropic,United States of America,API access,Closed weights, Gemini 3.1 Pro,Gemini 3.1 Pro,156.25,154.49,158.15,2026-02-19,Google DeepMind,United States of America,API access,Closed weights, Qwen3.8 Max (0902),Qwen3.8 Max (0902),155.98,153.89,158.27,2026-09-01,Alibaba,China,API access,Closed weights, DeepSeek V4 Pro 0813,DeepSeek V4 Pro 0813,155.96,154.1,157.99,2026-08-13,DeepSeek,China,Open weights (unrestricted),Open weights, Muse Spark Open,Muse Spark Open,155.45,153.37,157.73,2026-10-15,Meta AI,United States of America,Open weights (restricted use),Open weights, Gemini 3.5 Flash,Gemini 3.5 Flash,155.35,153.52,157.38,2026-05-19,Google DeepMind,United States of America,API access,Closed weights, Gemini 3.6 Flash,Gemini 3.6 Flash,155.24,153.42,157.18,2026-07-21,Google DeepMind,United States of America,API access,Closed weights, Muse Spark 1.2,Muse Spark 1.2,155.06,152.79,157.48,2026-08-05,Meta AI,United States of America,API access,Closed weights, Grok 4.5,Grok 4.5,154.97,153.23,156.91,2026-07-08,xAI,United States of America,API access,Closed weights, Muse Spark 1.1,Muse Spark 1.1,154.92,152.33,157.65,2026-07-09,Meta AI,United States of America,API access,Closed weights, Qwen3.7-Max,Qwen3.7-Max,154.58,152.57,156.67,2026-05-19,Alibaba,China,API access,Closed weights, Mistral Large 4,Mistral Large 4,154.56,151.79,157.42,2026-10-29,Mistral AI,France,Open weights (restricted use),Open weights, GPT-5.2,GPT-5.2,154.52,152.94,156.31,2025-12-11,OpenAI,United States of America,API access,Closed weights, DeepSeek V4 Flash 0731,DeepSeek V4 Flash 0731,154.43,152.55,156.42,2026-07-31,DeepSeek,China,Open weights (unrestricted),Open weights, GLM-5.3-Flash,GLM-5.3-Flash,153.78,152.04,155.62,2026-08-20,Z.ai (Zhipu AI),China,Open weights (unrestricted),Open weights, Muse Spark,Muse Spark,153.16,150.94,155.46,2026-04-08,Meta AI,United States of America,API access,Closed weights, GLM-5.2,GLM-5.2,153.04,151.35,154.86,2026-06-16,Z.ai (Zhipu AI),China,Open weights (unrestricted),Open weights, Gemini 3 Pro,Gemini 3 Pro,153.04,151.45,154.78,2025-11-18,Google DeepMind,United States of America,API access,Closed weights, Claude Sonnet 4.6,Claude Sonnet 4.6,152.86,151.17,154.66,2026-02-17,Anthropic,United States of America,API access,Closed weights, Mistral Large 4 Preview,Mistral Large 4 Preview,152.71,150.81,154.6,2026-10-06,Mistral AI,France,Open weights (restricted use),Open weights, Qwen 3.8 27B,Qwen 3.8 27B,152.65,150.7,154.76,2026-08-14,Alibaba,China,Open weights (unrestricted),Open weights, Gemini 3 Flash,Gemini 3 Flash,152.34,150.79,153.98,2025-12-17,Google DeepMind,United States of America,API access,Closed weights, Grok 4.20,Grok 4.20,152.3,150.58,154.09,2026-02-17,xAI,United States of America,API access,Closed weights, Claude Opus 4.5,Claude Opus 4.5,151.88,150.32,153.53,2025-11-24,Anthropic,United States of America,API access,Closed weights, Kimi K2.6,Kimi K2.6,151.62,149.86,153.44,2026-04-20,Moonshot,China,Open weights (unrestricted),Open weights, GPT-5 Pro,GPT-5 Pro,151.27,149.32,153.32,2025-10-07,OpenAI,United States of America,API access,Closed weights, Kimi K2.7 Code,Kimi K2.7 Code,151.15,149.18,153.25,2026-06-12,Moonshot,China,Open weights (unrestricted),Open weights, GPT-5.1,GPT-5.1,150.97,149.42,152.61,2025-11-13,OpenAI,United States of America,API access,Closed weights, Beam 1.0,Beam 1.0,150.87,148.66,153.12,2026-10-05,Reflection,United States of America,Open weights (unrestricted),Open weights, Inkling-Small,Inkling-Small,150.78,149.02,152.58,2026-07-15,Thinking Machines,United States of America,Open weights (unrestricted),Open weights, GLM-5.1,GLM-5.1,150.73,148.77,152.77,2026-04-07,Z.ai (Zhipu AI),China,Open weights (unrestricted),Open weights, DeepSeek-V4-Pro,DeepSeek-V4-Pro,150.21,148.47,152.04,2026-04-24,DeepSeek,China,Open weights (unrestricted),Open weights, GPT-5,GPT-5,150.0,,,2025-08-07,OpenAI,United States of America,API access,Closed weights, Grok 4.3 Beta,Grok 4.3 Beta,149.91,147.94,151.92,2026-04-17,xAI,United States of America,Hosted access (no API),Closed weights, Qwen 3.6 Max (Preview),Qwen 3.6 Max (Preview),149.88,147.87,152.07,2026-04-20,Alibaba,China,Unknown,Other, GPT-5.4 Mini,GPT-5.4 Mini,149.35,147.71,151.06,2026-03-17,OpenAI,United States of America,API access,Closed weights, Qwen 3.6 Plus,Qwen 3.6 Plus,148.91,146.96,150.92,2026-03-31,Alibaba,China,API access,Closed weights, Kimi K2.5,Kimi K2.5,148.8,147.3,150.35,2026-01-27,Moonshot,China,Open weights (unrestricted),Open weights, Inkling,Inkling,148.43,146.88,150.03,2026-07-15,Thinking Machines,United States of America,Open weights (unrestricted),Open weights, Qwen3.5 397B-A17B,Qwen3.5 397B-A17B,147.95,145.8,150.07,2026-02-13,Alibaba,China,Open weights (unrestricted),Open weights, Qwen3.7-Plus,Qwen3.7-Plus,147.95,145.91,150.04,2026-06-02,Alibaba,China,API access,Closed weights, MiniMax-M3,MiniMax-M3,147.9,145.76,150.04,2026-06-01,MiniMax,China,Open weights (restricted use),Open weights, Claude Sonnet 4.5,Claude Sonnet 4.5,147.59,146.14,149.04,2025-09-29,Anthropic,United States of America,API access,Closed weights, o3-pro,o3-pro,147.46,145.58,149.37,2025-06-10,OpenAI,United States of America,API access,Closed weights, Qwen 3.5 Plus (hosted 397B-A17B),Qwen 3.5 Plus (hosted 397B-A17B),147.3,145.38,149.2,2026-02-16,Alibaba,China,Unknown,Other, o3,o3,147.05,145.6,148.55,2025-04-16,OpenAI,United States of America,API access,Closed weights, Nemotron 3 Ultra,Nemotron 3 Ultra,147.0,145.02,149.09,2026-06-04,Nvidia,United States of America,Open weights (unrestricted),Open weights, Grok 4,Grok 4,146.96,145.48,148.46,2025-07-09,xAI,United States of America,API access,Closed weights, DeepSeek-V4-Flash,DeepSeek-V4-Flash,146.84,143.6,150.16,2026-04-24,DeepSeek,China,Open weights (unrestricted),Open weights, Qwen3.6 27B,Qwen3.6 27B,146.78,144.68,148.89,2026-04-22,Alibaba,China,Open weights (unrestricted),Open weights, GLM-5,GLM-5,146.54,144.97,148.11,2026-02-11,Z.ai (Zhipu AI),China,Open weights (unrestricted),Open weights, GPT-5.4 Nano,GPT-5.4 Nano,146.33,144.67,148.02,2026-03-17,OpenAI,United States of America,API access,Closed weights, DeepSeek-V3.2,DeepSeek-V3.2,146.26,144.75,147.75,2025-12-01,DeepSeek,China,Open weights (unrestricted),Open weights, GPT-5 mini,GPT-5 mini,146.01,144.4,147.59,2025-08-07,OpenAI,United States of America,API access,Closed weights, Gemini 3.5 Flash-Lite,Gemini 3.5 Flash-Lite,145.65,143.98,147.32,2026-07-21,Google DeepMind,United States of America,API access,Closed weights, Kimi K2 Thinking,Kimi K2 Thinking,145.48,143.69,147.27,2025-11-06,Moonshot,China,Open weights (restricted use),Open weights, o4-mini,o4-mini,145.22,143.68,146.7,2025-04-16,OpenAI,United States of America,API access,Closed weights, Gemini 3.1 Flash-Lite,Gemini 3.1 Flash-Lite,145.03,142.88,147.25,2026-03-03,Google,United States of America,API access,Closed weights, Gemini 2.5 Pro (Jun 2025),Gemini 2.5 Pro (Jun 2025),144.91,143.5,146.3,2025-06-05,Google DeepMind,United States of America,API access,Closed weights, Claude Opus 4.1,Claude Opus 4.1,144.77,143.0,146.5,2025-08-05,Anthropic,United States of America,API access,Closed weights, DeepSeek-V3.2-Exp,DeepSeek-V3.2-Exp,144.39,142.07,146.73,2025-09-29,DeepSeek,China,Open weights (unrestricted),Open weights, Grok 4 Fast,Grok 4 Fast,144.35,142.19,146.46,2025-09-19,xAI,United States of America,API access,Closed weights, Gemma 4 31B IT,Gemma 4 31B IT,144.12,141.72,146.59,2026-04-02,Google DeepMind,United States of America,Open weights (restricted use),Open weights, Qwen 3.6 35B-A3B,Qwen 3.6 35B-A3B,144.11,142.23,146.0,2026-04-14,Alibaba,China,Open weights (unrestricted),Open weights, Qwen3.5-35B-A3B,Qwen3.5-35B-A3B,144.08,141.59,146.7,2026-02-24,Alibaba,China,Open weights (unrestricted),Open weights, Gemini 2.5 Pro (Mar 2025),Gemini 2.5 Pro (Mar 2025),144.05,142.08,146.0,2025-03-31,Google DeepMind,United States of America,API access,Closed weights, Qwen 3.5 Flash (hosted 35B-A3B),Qwen 3.5 Flash (hosted 35B-A3B),143.87,142.01,145.75,2026-02-25,Alibaba,China,Unknown,Other, Qwen 3.6 Flash,Qwen 3.6 Flash,143.76,141.76,145.7,2026-04-27,Alibaba,China,Unknown,Other, GLM-4.7,GLM-4.7,143.36,141.4,145.27,2025-12-22,Z.ai (Zhipu AI),China,Open weights (unrestricted),Open weights, Qwen3-235B-A22B-Thinking (Jul 2025),Qwen3-235B-A22B-Thinking (Jul 2025),143.27,141.4,145.12,2025-07-25,Alibaba,China,Open weights (unrestricted),Open weights, Claude Opus 4,Claude Opus 4,142.88,141.27,144.47,2025-05-22,Anthropic,United States of America,API access,Closed weights, Claude Haiku 4.5,Claude Haiku 4.5,142.76,141.15,144.34,2025-10-15,Anthropic,United States of America,API access,Closed weights, Qwen3-Max,Qwen3-Max,142.44,140.29,144.54,2025-09-24,Alibaba,China,API access,Closed weights, Gemini 2.5 Flash (Sep 2025),Gemini 2.5 Flash (Sep 2025),141.94,139.1,144.59,2025-09-25,Google DeepMind,United States of America,API access,Closed weights, Claude Sonnet 4,Claude Sonnet 4,141.88,140.3,143.41,2025-05-22,Anthropic,United States of America,API access,Closed weights, Gemini 2.5 Flash (May 2025),Gemini 2.5 Flash (May 2025),141.58,139.44,143.64,2025-05-20,Google DeepMind,United States of America,API access,Closed weights, Gemma 4 26B A4B,Gemma 4 26B A4B,141.58,139.14,144.02,2026-04-02,Google DeepMind,United States of America,Open weights (unrestricted),Open weights, Mistral Medium 3.5,Mistral Medium 3.5,141.57,138.4,144.68,2026-04-28,Mistral AI,France,Open weights (restricted use),Open weights, Kimi K2 (Jul 2025),Kimi K2 (Jul 2025),141.25,138.25,143.95,2025-07-12,Moonshot,China,Open weights (restricted use),Open weights, o1,o1,141.18,139.48,142.79,2024-12-17,OpenAI,United States of America,API access,Closed weights, Claude 3.7 Sonnet,Claude 3.7 Sonnet,141.05,139.41,142.58,2025-02-24,Anthropic,United States of America,API access,Closed weights, Gemini 2.5 Flash (Jun 2025),Gemini 2.5 Flash (Jun 2025),140.84,138.4,143.12,2025-06-17,Google DeepMind,United States of America,API access,Closed weights, DeepSeek-V3.1,DeepSeek-V3.1,140.8,137.55,143.94,2025-08-21,DeepSeek,China,Open weights (unrestricted),Open weights, DeepSeek-R1 (May 2025),DeepSeek-R1 (May 2025),140.75,138.93,142.45,2025-05-28,DeepSeek,China,Open weights (unrestricted),Open weights, Gemini 2.5 Flash (Apr 2025),Gemini 2.5 Flash (Apr 2025),140.73,138.35,142.96,2025-04-17,Google DeepMind,United States of America,API access,Closed weights, o3-mini,o3-mini,140.65,138.96,142.25,2025-01-31,OpenAI,United States of America,API access,Closed weights, Gemini 2.5 Pro (May 2025),Gemini 2.5 Pro (May 2025),140.64,138.31,142.95,2025-05-06,Google DeepMind,United States of America,API access,Closed weights, gpt-oss-120b,gpt-oss-120b,140.23,138.26,142.08,2025-08-05,OpenAI,United States of America,Open weights (unrestricted),Open weights, Grok-3 mini,Grok-3 mini,139.95,138.15,141.69,2025-06-24,xAI,United States of America,API access,Closed weights, Qwen3.5-9B,Qwen3.5-9B,139.48,137.26,141.69,2026-02-24,Alibaba,China,Open weights (restricted use),Open weights, GPT-5 nano,GPT-5 nano,139.09,137.15,140.93,2025-08-07,OpenAI,United States of America,API access,Closed weights, DeepSeek-R1,DeepSeek-R1,138.63,136.7,140.4,2025-01-20,DeepSeek,China,Open weights (unrestricted),Open weights, Qwen3-235B-A22B,Qwen3-235B-A22B,138.43,136.11,140.67,2025-04-28,Alibaba,China,Open weights (unrestricted),Open weights, Qwen3-30B-A3B-Thinking (Jul 2025),Qwen3-30B-A3B-Thinking (Jul 2025),138.35,135.77,140.82,2025-07-30,Alibaba,China,Open weights (unrestricted),Open weights, Qwen3-235B-A22B-Instruct (Jul 2025),Qwen3-235B-A22B-Instruct (Jul 2025),138.07,135.66,140.24,2025-07-25,Alibaba,China,Open weights (unrestricted),Open weights, Grok 3,Grok 3,137.59,135.63,139.42,2025-04-09,xAI,United States of America,API access,Closed weights, Qwen3-32B,Qwen3-32B,136.84,134.41,139.22,2025-04-29,Alibaba,China,Open weights (unrestricted),Open weights, GPT-4.5,GPT-4.5,136.5,134.41,138.45,2025-02-27,OpenAI,United States of America,API access,Closed weights, QwQ-32B,QwQ-32B,136.48,133.72,139.15,2025-03-05,Alibaba,China,Open weights (unrestricted),Open weights, GPT-4.1,GPT-4.1,136.18,134.14,138.06,2025-04-14,OpenAI,United States of America,API access,Closed weights, Qwen3-14B,Qwen3-14B,136.08,133.42,138.55,2025-04-29,Alibaba,China,Open weights (unrestricted),Open weights, Gemini 2.0 Pro,Gemini 2.0 Pro,135.41,132.66,138.06,2025-02-05,Google DeepMind,United States of America,Hosted access (no API),Closed weights, DeepSeek-V3 (Mar 2025),DeepSeek-V3 (Mar 2025),135.28,133.12,137.34,2025-03-24,DeepSeek,China,Open weights (restricted use),Open weights, o1-mini,o1-mini,135.28,133.1,137.29,2024-09-12,OpenAI,United States of America,API access,Closed weights, gpt-oss-20b,gpt-oss-20b,134.88,132.27,137.31,2025-08-05,OpenAI,United States of America,Open weights (unrestricted),Open weights, Qwen3-30B-A3B-Instruct (Jul 2025),Qwen3-30B-A3B-Instruct (Jul 2025),134.83,131.99,137.52,2025-07-29,Alibaba,China,Open weights (unrestricted),Open weights, GPT-4.1 mini,GPT-4.1 mini,134.27,132.13,136.21,2025-04-14,OpenAI,United States of America,API access,Closed weights, Qwen3-30B-A3B,Qwen3-30B-A3B,134.18,131.45,136.76,2025-04-29,Alibaba,China,Open weights (unrestricted),Open weights, Gemini 2.0 Flash (Feb 2025),Gemini 2.0 Flash (Feb 2025),134.02,131.74,136.18,2025-02-05,"Google DeepMind,Google",United States of America,API access,Closed weights, Gemini 2.0 Flash Thinking (Jan 2025),Gemini 2.0 Flash Thinking (Jan 2025),133.61,130.54,136.4,2025-01-21,"Google DeepMind,Google",United States of America,API access,Closed weights, o1-preview,o1-preview,133.54,130.97,135.94,2024-09-12,OpenAI,United States of America,API access,Closed weights, Claude 3.5 Sonnet (October 2024),Claude 3.5 Sonnet (October 2024),133.34,131.04,135.54,2024-10-22,Anthropic,United States of America,API access,Closed weights, Gemini 2.5 Flash-Lite (Jun 2025),Gemini 2.5 Flash-Lite (Jun 2025),133.05,129.37,136.47,2025-06-17,Google DeepMind,United States of America,API access,Closed weights, Mistral Medium 3,Mistral Medium 3,132.81,130.17,135.2,2025-05-07,Mistral AI,France,API access,Closed weights, Llama 4 Maverick,Llama 4 Maverick,132.24,130.0,134.26,2025-04-06,Meta AI,United States of America,Open weights (restricted use),Open weights, Qwen3-8B,Qwen3-8B,132.08,129.06,134.78,2025-04-28,Alibaba,China,Open weights (unrestricted),Open weights, DeepSeek-V3,DeepSeek-V3,131.71,129.14,134.01,2024-12-26,DeepSeek,China,Open weights (restricted use),Open weights, Qwen2.5-Max,Qwen2.5-Max,131.17,128.42,133.62,2025-01-25,Alibaba,China,API access,Closed weights, Gemini 1.5 Pro (Sept 2024),Gemini 1.5 Pro (Sept 2024),130.45,127.81,132.84,2024-09-24,Google DeepMind,United States of America,API access,Closed weights, Claude 3.5 Sonnet,Claude 3.5 Sonnet,130.0,,,2024-06-20,Anthropic,United States of America,API access,Closed weights, Grok-2 (Dec 2024),Grok-2 (Dec 2024),129.75,127.01,132.25,2024-12-12,xAI,United States of America,API access,Closed weights, Phi-4,Phi-4,129.24,126.27,131.95,2024-12-12,Microsoft Research,United States of America,Open weights (unrestricted),Open weights, GPT-4o (Nov 2024),GPT-4o (Nov 2024),128.78,126.12,131.24,2024-11-20,OpenAI,United States of America,API access,Closed weights, GPT-4o (May 2024),GPT-4o (May 2024),128.13,125.37,130.68,2024-05-13,OpenAI,United States of America,API access,Closed weights, Llama 3.1-405B,Llama 3.1-405B,127.87,124.95,130.51,2024-07-23,Meta AI,United States of America,Open weights (restricted use),Open weights, Llama 4 Scout,Llama 4 Scout,127.77,124.88,130.4,2025-04-05,Meta AI,United States of America,Open weights (restricted use),Open weights, Gemma 3 27B,Gemma 3 27B,127.5,124.41,130.28,2025-03-12,Google DeepMind,United States of America,Open weights (restricted use),Open weights, Qwen2.5-72B,Qwen2.5-72B,127.21,124.47,129.8,2024-09-19,Alibaba,China,Open weights (unrestricted),Open weights, GPT-4o (Aug 2024),GPT-4o (Aug 2024),127.12,124.07,129.86,2024-08-06,OpenAI,United States of America,API access,Closed weights, Gemini 1.5 Flash (Sep 2024),Gemini 1.5 Flash (Sep 2024),126.96,123.87,129.8,2024-09-24,Google DeepMind,United States of America,API access,Closed weights, GPT-4.1 nano,GPT-4.1 nano,126.81,123.79,129.59,2025-04-14,OpenAI,United States of America,API access,Closed weights, Mistral Large 2 (Jul 2024),Mistral Large 2 (Jul 2024),126.42,123.24,129.4,2024-07-24,Mistral AI,France,Open weights (non-commercial),Open weights, Llama 3.3 70B,Llama 3.3 70B,126.16,123.1,128.98,2024-12-06,Meta AI,United States of America,Open weights (restricted use),Open weights, GPT-4 Turbo (Apr 2024),GPT-4 Turbo (Apr 2024),125.82,122.84,128.62,2024-04-09,OpenAI,United States of America,API access,Closed weights, Claude 3 Opus,Claude 3 Opus,125.58,122.65,128.17,2024-02-29,Anthropic,United States of America,API access,Closed weights, Gemini 1.5 Pro (May 2024),Gemini 1.5 Pro (May 2024),125.37,122.09,128.37,2024-05-14,Google DeepMind,United States of America,API access,Closed weights, Claude 3.5 Haiku,Claude 3.5 Haiku,124.75,121.49,127.65,2024-10-22,Anthropic,United States of America,API access,Closed weights, GPT-4 Turbo (Nov 2023),GPT-4 Turbo (Nov 2023),124.75,121.47,127.68,2024-01-25,OpenAI,United States of America,API access,Closed weights, Llama 3.1-70B,Llama 3.1-70B,124.72,121.58,127.63,2024-07-23,Meta AI,United States of America,Open weights (restricted use),Open weights, GPT-4o mini,GPT-4o mini,123.75,120.5,126.65,2024-07-18,OpenAI,United States of America,API access,Closed weights, Llama 3.2 90B,Llama 3.2 90B,122.89,119.04,126.23,2024-09-24,Meta AI,United States of America,Open weights (restricted use),Open weights, Qwen2-72B,Qwen2-72B,122.33,118.73,125.52,2024-06-07,Alibaba,China,Open weights (unrestricted),Open weights, GPT-4 (Mar 2023),GPT-4 (Mar 2023),121.58,117.72,125.02,2023-03-14,OpenAI,United States of America,API access,Closed weights, Llama 3-70B,Llama 3-70B,121.34,117.67,124.75,2024-04-18,Meta AI,United States of America,Open weights (restricted use),Open weights, "DeepSeek-V2 (MoE-236B, May 2024)","DeepSeek-V2 (MoE-236B, May 2024)",120.76,113.88,127.62,2024-05-07,DeepSeek,China,Open weights (restricted use),Open weights, Gemini 1.5 Flash (May 2024),Gemini 1.5 Flash (May 2024),120.55,116.51,124.09,2024-05-23,Google DeepMind,United States of America,API access,Closed weights, Mistral Large,Mistral Large,119.96,115.99,123.45,2024-02-26,Mistral AI,France,API access,Closed weights, GPT-4 (Jun 2023),GPT-4 (Jun 2023),119.82,115.94,123.33,2023-06-13,OpenAI,United States of America,API access,Closed weights, Claude 3 Sonnet,Claude 3 Sonnet,119.65,115.71,123.23,2024-02-29,Anthropic,United States of America,API access,Closed weights, Qwen2.5-Coder-32B,Qwen2.5-Coder-32B,118.74,110.04,127.38,2024-09-18,Alibaba,China,Open weights (unrestricted),Open weights, Mixtral 8x22B,Mixtral 8x22B,118.66,114.34,122.54,2024-04-17,Mistral AI,France,Open weights (unrestricted),Open weights, Gemma 2 27B,Gemma 2 27B,118.61,114.48,122.35,2024-06-24,Google DeepMind,United States of America,Open weights (restricted use),Open weights, Claude 3 Haiku,Claude 3 Haiku,117.79,113.6,121.58,2024-03-07,Anthropic,United States of America,API access,Closed weights, Claude 2,Claude 2,117.12,112.61,121.05,2023-07-11,Anthropic,United States of America,API access,Closed weights, phi-3-small 7.4B,phi-3-small 7.4B,116.26,110.26,121.84,2024-04-23,Microsoft,United States of America,Open weights (unrestricted),Open weights, phi-3-medium 14B,phi-3-medium 14B,114.01,109.14,118.32,2024-04-23,Microsoft,United States of America,Open weights (unrestricted),Open weights, Claude Instant,Claude Instant,114.0,107.19,120.14,2023-08-09,Anthropic,United States of America,API access,Closed weights, Mistral NeMo,Mistral NeMo,113.81,108.8,118.19,2024-07-18,Mistral AI,France,Open weights (unrestricted),Open weights, Qwen2.5-Coder-14B,Qwen2.5-Coder-14B,113.72,104.82,122.09,2024-09-18,,,Unknown,Other, PaLM 2-L,PaLM 2-L,113.71,98.69,129.5,2023-05-17,,,Unknown,Other, Mixtral 8x7B,Mixtral 8x7B,113.1,108.13,117.48,2023-12-11,Mistral AI,France,Open weights (unrestricted),Open weights, GPT-3.5 Turbo (Jan 2024),GPT-3.5 Turbo (Jan 2024),112.77,107.84,117.15,2024-01-25,OpenAI,United States of America,API access,Closed weights, GPT-3.5 Turbo (Nov 2023),GPT-3.5 Turbo (Nov 2023),112.57,107.25,117.26,2023-11-06,OpenAI,United States of America,API access,Closed weights, Gemma 2 9B,Gemma 2 9B,112.28,106.66,117.21,2024-06-24,Google DeepMind,United States of America,Open weights (restricted use),Open weights, Llama 3.1-8B,Llama 3.1-8B,110.78,105.32,115.82,2024-07-23,Meta AI,United States of America,Open weights (restricted use),Open weights, phi-3-mini 3.8B,phi-3-mini 3.8B,110.4,104.13,116.07,2024-04-23,Microsoft,United States of America,Open weights (unrestricted),Open weights, Stable Beluga 2,Stable Beluga 2,109.29,101.63,115.99,2023-07-20,Stability AI,United Kingdom,Open weights (non-commercial),Open weights, Llama 3-8B,Llama 3-8B,109.0,103.32,113.95,2024-04-18,Meta AI,United States of America,Open weights (restricted use),Open weights, Qwen2.5-Coder (7B),Qwen2.5-Coder (7B),107.45,98.13,115.86,2024-09-18,Alibaba,China,Open weights (unrestricted),Open weights, Falcon-180B,Falcon-180B,106.92,98.04,115.21,2023-09-06,Technology Innovation Institute,United Arab Emirates,Open weights (restricted use),Open weights, GPT-3.5 Turbo (Jun 2023),GPT-3.5 Turbo (Jun 2023),105.84,97.72,113.26,2023-06-13,OpenAI,United States of America,API access,Closed weights, Llama 2-70B,Llama 2-70B,105.64,99.62,110.86,2023-07-18,Meta AI,United States of America,Open weights (restricted use),Open weights, Gemma 7B,Gemma 7B,102.87,94.96,109.59,2024-02-21,Google DeepMind,United States of America,Open weights (restricted use),Open weights, Mistral 7B v0.1,Mistral 7B v0.1,102.56,94.84,109.26,2023-09-27,Mistral AI,France,Open weights (unrestricted),Open weights, Yi-9B,Yi-9B,102.06,90.22,112.03,2024-03-01,,,Unknown,Other, Qwen-14B,Qwen-14B,101.61,92.75,109.54,2023-09-24,Alibaba,China,Open weights (restricted use),Open weights, LLaMA-65B,LLaMA-65B,101.38,92.98,108.66,2023-02-24,Meta AI,United States of America,Open weights (non-commercial),Open weights, DeepSeek-Coder-V2-Lite-Base,DeepSeek-Coder-V2-Lite-Base,99.31,88.78,108.58,2024-06-13,,,Unknown,Other, Falcon 2 11B,Falcon 2 11B,99.01,87.9,109.09,2024-05-09,Technology Innovation Institute,United Arab Emirates,Open weights (restricted use),Open weights, Nemotron-4 15B,Nemotron-4 15B,98.97,89.56,106.97,2024-02-26,Nvidia,United States of America,Unreleased,Closed weights, internlm-20b,internlm-20b,98.79,88.76,107.58,2023-09-18,,,Unknown,Other, PaLM 2-M,PaLM 2-M,97.77,84.79,109.73,2023-05-17,,,Unknown,Other, Qwen2.5-Coder-3B,Qwen2.5-Coder-3B,96.63,82.05,108.97,2024-09-18,,,Unknown,Other, LLaMA-33B,LLaMA-33B,96.37,87.05,104.65,2023-02-24,Meta AI,United States of America,Open weights (non-commercial),Open weights, Llama 2-13B,Llama 2-13B,96.28,87.19,104.31,2023-07-18,Meta AI,United States of America,Open weights (restricted use),Open weights, StarCoder 2 15B,StarCoder 2 15B,96.04,84.83,105.99,2024-02-20,"Hugging Face,ServiceNow,Nvidia,BigCode",United States of America,Open weights (restricted use),Open weights, Phi-2,Phi-2,95.94,86.5,104.28,2023-12-12,Microsoft,United States of America,Open weights (unrestricted),Open weights, Yi 6B,Yi 6B,94.06,83.69,102.91,2023-11-22,01.AI,China,Open weights (restricted use),Open weights, Llama 2-34B,Llama 2-34B,93.69,83.73,102.35,2023-07-18,Meta AI,United States of America,Unreleased,Closed weights, PaLM 2-S,PaLM 2-S,93.62,80.24,105.75,2023-05-17,,,Unknown,Other, Baichuan2-13B,Baichuan2-13B,91.59,80.91,100.91,2023-09-06,Baichuan,China,Open weights (restricted use),Open weights, Falcon-40B,Falcon-40B,90.42,79.63,99.75,2023-05-25,Technology Innovation Institute,United Arab Emirates,Open weights (unrestricted),Open weights, Qwen2.5-Coder (1.5B),Qwen2.5-Coder (1.5B),89.01,76.06,99.99,2024-09-18,Alibaba,China,Open weights (unrestricted),Open weights, Qwen-7B,Qwen-7B,88.87,77.54,98.62,2023-09-28,Alibaba,China,Open weights (restricted use),Open weights, MPT-30B,MPT-30B,84.95,72.95,95.48,2023-06-22,MosaicML,United States of America,Open weights (unrestricted),Open weights, LLaMA-13B,LLaMA-13B,84.95,72.99,95.23,2023-02-24,Meta AI,United States of America,Open weights (non-commercial),Open weights, internlm-7b,internlm-7b,84.31,72.08,94.94,2023-07-05,,,Unknown,Other, Baichuan 2-7B,Baichuan 2-7B,83.25,70.59,94.26,2023-09-20,Baichuan,China,Open weights (restricted use),Open weights, INTELLECT-1,INTELLECT-1,83.09,70.64,93.9,2024-11-29,"Prime Intellect,Hugging Face,Arcee AI",United States of America,Unknown,Other, Llama 2-7B,Llama 2-7B,82.8,70.34,93.04,2023-07-18,Meta AI,United States of America,Open weights (restricted use),Open weights, LLaMA-7B,LLaMA-7B,76.96,62.98,88.78,2023-02-24,Meta AI,United States of America,Open weights (non-commercial),Open weights, DeepSeek Coder 33B,DeepSeek Coder 33B,75.93,59.8,89.43,2023-11-02,"DeepSeek,Peking University",China,Open weights (restricted use),Open weights, vicuna-13b-v1.1,vicuna-13b-v1.1,75.78,60.52,88.45,2023-04-12,,,Unknown,Other, Gemma 2B,Gemma 2B,75.36,60.81,87.85,2024-02-21,Google DeepMind,United States of America,Open weights (restricted use),Open weights, CodeQwen1.5-7B,CodeQwen1.5-7B,74.53,58.12,88.3,2024-04-15,,,Unknown,Other, chatglm2-6b,chatglm2-6b,73.4,56.39,86.83,2023-06-24,,,Unknown,Other, MPT-7B,MPT-7B,73.18,58.58,86.14,2023-05-05,MosaicML,United States of America,Open weights (unrestricted),Open weights, Falcon-7B,Falcon-7B,72.84,57.82,85.82,2023-04-24,Technology Innovation Institute,United Arab Emirates,Open weights (unrestricted),Open weights, StarCoder 2 7B,StarCoder 2 7B,72.14,54.51,86.47,2024-02-20,"Hugging Face,ServiceNow,Nvidia,BigCode",United States of America,Open weights (restricted use),Open weights, XGen-7B,XGen-7B,69.95,53.38,83.98,2023-06-27,Salesforce,United States of America,Open weights (unrestricted),Open weights, Phi-1.5,Phi-1.5,68.17,49.93,83.48,2023-09-11,Microsoft,United States of America,Open weights (unrestricted),Open weights, Qwen2.5-Coder-0.5B,Qwen2.5-Coder-0.5B,67.95,49.04,83.62,2024-09-18,,,Unknown,Other, DeepSeek Coder 6.7B,DeepSeek Coder 6.7B,67.02,48.31,82.73,2023-11-02,"DeepSeek,Peking University",China,Open weights (restricted use),Open weights, StarCoder 2 3B,StarCoder 2 3B,66.21,47.86,81.82,2024-02-22,"Hugging Face,ServiceNow,Nvidia,BigCode",United States of America,Open weights (restricted use),Open weights, open_llama_7b,open_llama_7b,63.91,45.99,78.99,2023-06-07,,,Unknown,Other, Qwen-1_8B,Qwen-1_8B,59.23,40.43,75.43,2023-11-30,,,Unknown,Other, RedPajama-INCITE-7B-Base,RedPajama-INCITE-7B-Base,57.17,37.38,74.28,2023-05-04,,,Unknown,Other, Dolly 2.0-12b,Dolly 2.0-12b,55.1,34.71,72.43,2023-04-11,Databricks,United States of America,Open weights (unrestricted),Open weights, Cerebras-GPT-13B,Cerebras-GPT-13B,48.59,25.79,67.73,2023-03-20,Cerebras Systems,United States of America,Open weights (unrestricted),Open weights,