Ranking de modelos de IA
¿Qué modelo de IA es el más capaz? Rankings basados en pruebas independientes realizadas por Epoch AI, junto a los precios, para que encuentres el mejor modelo para tu presupuesto.
Selección rápida
| # | Modelo | Índice generali | GPQA Diamondi | SWE-bench Verifiedi | Entradai | Salidai | Lanzamiento |
|---|---|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 166.6 | 95.8% | — | 10 US$ | 50 US$ | sept 2026 |
| 2 | Claude Fable 5.1Anthropic | 165.0 | — | — | 10 US$ | 50 US$ | sept 2026 |
| 3 | Claude Fable 5Anthropic | 163.6 | 85.9% | — | 10 US$ | 50 US$ | jun 2026 |
| 4 | Claude Opus 5Anthropic | 162.7 | 93.9% | — | 5 US$ | 25 US$ | jul 2026 |
| 5 | GPT-5.5 ProOpenAI | 162.5 | — | — | 30 US$ | 180 US$ | abr 2026 |
| 6 | GPT-5.6 SolOpenAI | 162.0 | 93.5% | — | 2 US$ | 10 US$ | jul 2026 |
| 7 | GPT-5.6 TerraOpenAI | 159.3 | 93.3% | — | 2 US$ | 12 US$ | jul 2026 |
| 8 | GPT-5.5OpenAI | 159.3 | 90.7% | — | 5 US$ | 30 US$ | abr 2026 |
| 9 | GPT-5.4 ProOpenAI | 159.1 | 94.6% | — | 30 US$ | 180 US$ | mar 2026 |
| 10 | Claude Opus 4.8Anthropic | 158.3 | 91.0% | — | 5 US$ | 25 US$ | may 2026 |
| 11 | Gemini 3.7 FlashGoogle DeepMind | 157.7 | 94.8% | — | 0,75 US$ | 3,75 US$ | ago 2026 |
| 12 | Kimi K3Moonshot AI | 157.7 | 93.1% | — | 3 US$ | 15 US$ | jul 2026 |
| 13 | Gemini 3.8 FlashGoogle DeepMind | 157.1 | 95.4% | — | 0,75 US$ | 3,75 US$ | sept 2026 |
| 14 | Muse Spark 1.3Meta | 156.9 | — | — | 1,25 US$ | 4,25 US$ | sept 2026 |
| 15 | GPT-5.4OpenAI | 156.9 | 93.3% | 76.9% | 2,50 US$ | 15 US$ | mar 2026 |
| 16 | GPT-5.3-CodexOpenAI | 156.8 | — | 74.8% | 1,75 US$ | 14 US$ | feb 2026 |
| 17 | Qwen3.8 Max (0902)Alibaba Qwen | 156.7 | 92.7% | — | 2 US$ | 6 US$ | sept 2026 |
| 18 | Grok 4.6xAI | 156.5 | 94.0% | — | 2 US$ | 6 US$ | ago 2026 |
| 19 | Claude Opus 4.7Anthropic | 156.4 | 90.2% | 83.5% | 5 US$ | 25 US$ | abr 2026 |
| 20 | GPT-5.6 LunaOpenAI | 156.3 | 91.6% | — | 0,20 US$ | 1,20 US$ | jul 2026 |
| 21 | Claude Sonnet 5Anthropic | 156.3 | 90.5% | — | 2 US$ | 10 US$ | jun 2026 |
| 22 | GLM 5.3Z.ai (Zhipu) | 155.6 | 90.9% | — | 1,40 US$ | 4,40 US$ | ago 2026 |
| 23 | DeepSeek V4 Pro 0813DeepSeek | 155.4 | 91.7% | 77.6% | 0,66 US$ | 1,98 US$ | ago 2026 |
| 24 | Claude Opus 4.6Anthropic | 155.4 | 90.5% | 78.7% | 5 US$ | 25 US$ | feb 2026 |
| 25 | GPT-5.2 ProOpenAI | 155.4 | — | — | 21 US$ | 168 US$ | dic 2025 |
| 26 | Muse Spark 1.2Meta | 155.2 | — | — | 1,25 US$ | 4,25 US$ | ago 2026 |
| 27 | DeepSeek V4.1 FlashDeepSeek | 155.0 | — | — | 0,30 US$ | 1,20 US$ | sept 2026 |
| 28 | Gemini 3.1 Pro PreviewGoogle DeepMind | 154.9 | 94.4% | — | 2 US$ | 12 US$ | feb 2026 |
| 29 | Gemini 3.5 FlashGoogle DeepMind | 154.6 | 92.8% | 79.3% | 1,50 US$ | 9 US$ | may 2026 |
| 30 | DeepSeek V4 Flash 0731DeepSeek | 154.5 | 91.0% | — | 0,018 US$ | 0,32 US$ | jul 2026 |
| 31 | Gemini 3.6 FlashGoogle DeepMind | 154.4 | 94.1% | — | 0,75 US$ | 3,75 US$ | jul 2026 |
| 32 | Muse Spark 1.1Meta | 154.3 | — | — | 1,25 US$ | 4,25 US$ | jul 2026 |
| 33 | Grok 4.5xAI | 154.0 | 93.4% | — | 2 US$ | 6 US$ | jul 2026 |
| 34 | Qwen3.7 MaxAlibaba Qwen | 153.7 | 90.9% | 77.3% | 1,48 US$ | 4,43 US$ | may 2026 |
| 35 | GPT-5.2 ChatOpenAI | 153.5 | 91.4% | 73.8% | 1,75 US$ | 14 US$ | dic 2025 |
| 36 | Claude Sonnet 4.6Anthropic | 152.3 | 87.4% | 75.2% | 3 US$ | 15 US$ | feb 2026 |
| 37 | Grok 4.20xAI | 152.0 | — | — | 1,25 US$ | 2,50 US$ | mar 2026 |
| 38 | GLM 5.3 FlashZ.ai (Zhipu) | 151.9 | 90.2% | — | 0,15 US$ | 0,50 US$ | ago 2026 |
| 39 | GLM 5.2Z.ai (Zhipu) | 151.8 | 91.9% | 78.7% | 0,14 US$ | 3,99 US$ | jun 2026 |
| 40 | Gemini 3 Flash PreviewGoogle DeepMind | 151.8 | 89.4% | 75.4% | 0,50 US$ | 3 US$ | dic 2025 |
| 41 | Kimi K2.6Moonshot AI | 151.0 | 90.8% | 76.7% | 0,65 US$ | 3,41 US$ | abr 2026 |
| 42 | GPT-5 ProOpenAI | 150.3 | — | — | 15 US$ | 120 US$ | oct 2025 |
| 43 | Inkling SmallThinking Machines Lab | 150.1 | 88.5% | — | 0,45 US$ | 1,20 US$ | jul 2026 |
| 44 | Claude Opus 4.5Anthropic | 150.1 | 86.0% | 76.7% | 5 US$ | 25 US$ | nov 2025 |
| 45 | Kimi K2.7 CodeMoonshot AI | 150.0 | 87.9% | — | 0,66 US$ | 3,30 US$ | jun 2026 |
| 46 | GPT-5OpenAI | 150.0 | 86.2% | 73.6% | 1,25 US$ | 10 US$ | ago 2025 |
| 47 | GLM 5.1Z.ai (Zhipu) | 149.9 | 89.9% | 74.2% | 1,40 US$ | 4,40 US$ | abr 2026 |
| 48 | GPT-5.1OpenAI | 149.7 | 87.6% | 68.0% | 1,25 US$ | 10 US$ | nov 2025 |
| 49 | Qwen3.8 27BAlibaba Qwen | 149.4 | — | — | 0,42 US$ | 3 US$ | ago 2026 |
| 50 | Qwen3.6 Max PreviewAlibaba Qwen | 149.3 | 87.4% | 76.7% | 1,03 US$ | 6,16 US$ | abr 2026 |
| 51 | GPT-5.4 MiniOpenAI | 148.8 | 86.9% | — | 0,75 US$ | 4,50 US$ | mar 2026 |
| 52 | InklingThinking Machines Lab | 148.6 | 88.3% | — | 1 US$ | 4,05 US$ | jul 2026 |
| 53 | Kimi K2.5Moonshot AI | 148.1 | — | 73.8% | 0,45 US$ | 2,25 US$ | ene 2026 |
| 54 | Qwen3.6 PlusAlibaba Qwen | 147.7 | 88.4% | 57.9% | 0,33 US$ | 1,95 US$ | abr 2026 |
| 55 | Qwen3.7 PlusAlibaba Qwen | 147.4 | 87.9% | — | 0,32 US$ | 1,28 US$ | jun 2026 |
| 56 | o3 ProOpenAI | 147.4 | — | — | 20 US$ | 80 US$ | jun 2025 |
| 57 | MiniMax M3MiniMax | 147.0 | 90.9% | — | 0,30 US$ | 1,20 US$ | may 2026 |
| 58 | o3OpenAI | 146.9 | 81.8% | 62.3% | 2 US$ | 8 US$ | abr 2025 |
| 59 | Qwen3.5 Plus 2026-04-20Alibaba Qwen | 146.8 | 84.8% | — | 0,30 US$ | 1,80 US$ | abr 2026 |
| 60 | Claude Sonnet 4.5Anthropic | 146.8 | 82.3% | 71.3% | 3 US$ | 15 US$ | sept 2025 |
| 61 | MiniMax M2.5MiniMax | 146.7 | — | — | 0,27 US$ | 1,08 US$ | feb 2026 |
| 62 | Qwen3.5 397B A17BAlibaba Qwen | 146.6 | 86.4% | — | 0,55 US$ | 3,50 US$ | feb 2026 |
| 63 | Qwen3.6 27BAlibaba Qwen | 146.5 | 85.9% | — | 0,32 US$ | 3,20 US$ | abr 2026 |
| 64 | DeepSeek V3.2DeepSeek | 146.3 | — | — | 0,28 US$ | 0,42 US$ | dic 2025 |
| 65 | Nemotron 3 UltraNVIDIA | 146.2 | 85.4% | — | 0,60 US$ | 2,40 US$ | jun 2026 |
| 66 | Kimi K2 ThinkingMoonshot AI | 146.0 | — | — | 0,60 US$ | 2,50 US$ | nov 2025 |
| 67 | MiniMax M2.7MiniMax | 145.8 | — | — | 0,21 US$ | 0,84 US$ | mar 2026 |
| 68 | GPT-5.4 NanoOpenAI | 145.8 | 78.5% | — | 0,20 US$ | 1,25 US$ | mar 2026 |
| 69 | GLM 5Z.ai (Zhipu) | 145.8 | 87.8% | 72.1% | 0,60 US$ | 1,92 US$ | feb 2026 |
| 70 | o4 MiniOpenAI | 145.6 | 79.6% | — | 1,10 US$ | 4,40 US$ | abr 2025 |
| 71 | GPT-5 MiniOpenAI | 145.5 | 75.0% | 64.7% | 0,25 US$ | 2 US$ | ago 2025 |
| 72 | Gemini 2.5 ProGoogle DeepMind | 145.3 | 85.3% | 57.6% | 1,25 US$ | 10 US$ | jun 2025 |
| 73 | Gemini 3.5 Flash LiteGoogle DeepMind | 145.1 | 83.3% | — | 0,30 US$ | 2,50 US$ | jul 2026 |
| 74 | DeepSeek V3.2 ExpDeepSeek | 145.0 | — | — | 0,27 US$ | 0,41 US$ | sept 2025 |
| 75 | Qwen3.7 FlashAlibaba Qwen | 144.6 | 82.3% | — | 0,03 US$ | 0,13 US$ | jul 2026 |
| 76 | Gemini 3.1 Flash LiteGoogle DeepMind | 144.5 | 81.8% | — | 0,25 US$ | 1,50 US$ | may 2026 |
| 77 | Claude Opus 4.1Anthropic | 144.1 | 77.3% | 73.3% | 15 US$ | 75 US$ | ago 2025 |
| 78 | Qwen3.5-FlashAlibaba Qwen | 144.0 | 82.3% | — | 0,065 US$ | 0,26 US$ | feb 2026 |
| 79 | Qwen3.6 35B A3BAlibaba Qwen | 143.9 | 84.8% | — | 0,15 US$ | 1 US$ | abr 2026 |
| 80 | Qwen3 235B A22B Thinking 2507Alibaba Qwen | 143.9 | 80.1% | — | 0,23 US$ | 2,30 US$ | jul 2025 |
| 81 | GLM 4.7Z.ai (Zhipu) | 143.5 | 83.3% | — | 0,60 US$ | 2,20 US$ | dic 2025 |
| 82 | Qwen3.6 FlashAlibaba Qwen | 143.3 | 83.3% | — | 0,19 US$ | 1,13 US$ | abr 2026 |
| 83 | Gemini 2.5 FlashGoogle DeepMind | 143.0 | — | — | 0,30 US$ | 2,50 US$ | jun 2025 |
| 84 | Gemma 4 31BGoogle DeepMind | 142.7 | 75.8% | — | 0,09 US$ | 0,34 US$ | abr 2026 |
| 85 | Qwen3.5-35B-A3BAlibaba Qwen | 142.5 | 83.5% | — | 0,16 US$ | 1,30 US$ | feb 2026 |
| 86 | Claude Haiku 4.5Anthropic | 142.4 | 71.2% | — | 1 US$ | 5 US$ | oct 2025 |
| 87 | Qwen3 MaxAlibaba Qwen | 142.4 | 72.6% | — | 0,78 US$ | 3,90 US$ | sept 2025 |
| 88 | Gemma 4 26B A4B Google DeepMind | 141.9 | 73.2% | — | 0,077 US$ | 0,26 US$ | abr 2026 |
| 89 | o1OpenAI | 141.9 | 76.8% | — | 15 US$ | 60 US$ | dic 2024 |
| 90 | Claude Sonnet 4Anthropic | 141.7 | 79.2% | — | 3 US$ | 15 US$ | may 2025 |
| 91 | Mistral Medium 3.5Mistral AI | 141.4 | — | — | 1,50 US$ | 7,50 US$ | abr 2026 |
| 92 | R1 0528DeepSeek | 141.3 | 76.3% | — | 0,50 US$ | 2,15 US$ | may 2025 |
| 93 | o3 MiniOpenAI | 140.4 | 77.0% | — | 1,10 US$ | 4,40 US$ | ene 2025 |
| 94 | Kimi K2 0905Moonshot AI | 140.1 | — | — | 0,60 US$ | 2,50 US$ | sept 2025 |
| 95 | DeepSeek V3.1DeepSeek | 139.9 | — | — | 0,25 US$ | 0,95 US$ | ago 2025 |
| 96 | gpt-oss-120bOpenAI | 139.9 | 75.8% | — | 0,037 US$ | 0,17 US$ | ago 2025 |
| 97 | Qwen3 30B A3B Thinking 2507Alibaba Qwen | 139.6 | 70.1% | — | 0,20 US$ | 2,40 US$ | ago 2025 |
| 98 | Qwen3.5-9BAlibaba Qwen | 139.5 | 79.0% | — | 0,10 US$ | 0,15 US$ | mar 2026 |
| 99 | GPT-5 NanoOpenAI | 139.4 | 69.4% | — | 0,05 US$ | 0,40 US$ | ago 2025 |
| 100 | Qwen3 235B A22B Instruct 2507Alibaba Qwen | 139.4 | 70.7% | — | 0,088 US$ | 0,35 US$ | jul 2025 |
| 101 | Qwen3 32BAlibaba Qwen | 138.5 | 65.7% | — | 0,08 US$ | 0,28 US$ | abr 2025 |
| 102 | Qwen3 14BAlibaba Qwen | 138.2 | 63.8% | — | 0,12 US$ | 0,24 US$ | abr 2025 |
| 103 | gpt-oss-20bOpenAI | 137.8 | 60.8% | — | 0,018 US$ | 0,09 US$ | ago 2025 |
| 104 | Qwen3 30B A3B Instruct 2507Alibaba Qwen | 137.4 | 61.7% | — | 0,048 US$ | 0,19 US$ | jul 2025 |
| 105 | GPT-4.1OpenAI | 136.8 | 66.9% | 48.5% | 2 US$ | 8 US$ | abr 2025 |
| 106 | Qwen3 8BAlibaba Qwen | 136.2 | 56.8% | — | 0,12 US$ | 0,46 US$ | abr 2025 |
| 107 | DeepSeek V3DeepSeek | 136.0 | 71.2% | — | 0,26 US$ | 1,03 US$ | dic 2024 |
| 108 | GPT-4.1 MiniOpenAI | 135.0 | 65.8% | — | 0,40 US$ | 1,60 US$ | abr 2025 |
| 109 | Mistral Medium 3Mistral AI | 134.1 | — | — | 0,40 US$ | 2 US$ | may 2025 |
| 110 | Gemini 2.5 Flash LiteGoogle DeepMind | 133.9 | — | — | 0,10 US$ | 0,40 US$ | jul 2025 |
| 111 | Llama 4 MaverickMeta | 132.2 | — | — | 0,19 US$ | 0,65 US$ | abr 2025 |
| 112 | Phi 4Microsoft | 130.4 | 56.1% | — | 0,07 US$ | 0,14 US$ | ene 2025 |
| 113 | Gemma 3 27BGoogle DeepMind | 130.0 | 47.7% | — | 0,08 US$ | 0,45 US$ | mar 2025 |
| 114 | GPT-4.1 NanoOpenAI | 129.6 | 48.9% | — | 0,10 US$ | 0,40 US$ | abr 2025 |
| 115 | Llama 4 ScoutMeta | 129.6 | — | — | 0,10 US$ | 0,30 US$ | abr 2025 |
| 116 | GPT-4o (2024-11-20)OpenAI | 129.0 | 49.2% | 31.0% | 2,50 US$ | 10 US$ | nov 2024 |
| 117 | Qwen2.5 72B InstructAlibaba Qwen | 129.0 | 49.1% | — | 0,36 US$ | 0,40 US$ | sept 2024 |
| 118 | 127.3 | 47.4% | — | 0,10 US$ | 0,32 US$ | dic 2024 | |
| 119 | GPT-4 TurboOpenAI | 127.3 | 46.6% | — | 10 US$ | 30 US$ | abr 2024 |
| 120 | Mistral Small 3Mistral AI | 127.1 | 47.3% | — | 0,05 US$ | 0,08 US$ | ene 2025 |
| 121 | GPT-4o-miniOpenAI | 126.6 | 37.7% | — | 0,15 US$ | 0,60 US$ | jul 2024 |
| 122 | 125.9 | 44.2% | — | 0,40 US$ | 0,40 US$ | jul 2024 | |
| 123 | GPT-4OpenAI | 125.9 | 35.7% | — | 30 US$ | 60 US$ | may 2023 |
| 124 | Gemma 3 12BGoogle DeepMind | 123.5 | 39.5% | — | 0,05 US$ | 0,15 US$ | mar 2025 |
| 125 | Gemma 2 27BGoogle DeepMind | 122.1 | 36.5% | — | 0,65 US$ | 0,65 US$ | jul 2024 |
| 126 | Mistral Large 3 2512Mistral AI | 122.0 | 51.3% | — | 0,50 US$ | 1,50 US$ | dic 2025 |
| 127 | Mixtral 8x22B InstructMistral AI | 122.0 | — | — | 2 US$ | 6 US$ | abr 2024 |
| 128 | Qwen2.5 Coder 32B InstructAlibaba Qwen | 119.4 | — | — | 0,66 US$ | 1 US$ | nov 2024 |
| 129 | Command R (08-2024)Cohere | 119.3 | — | — | 0,15 US$ | 0,60 US$ | ago 2024 |
| 130 | Mistral NemoMistral AI | 118.6 | — | — | 0,019 US$ | 0,03 US$ | jul 2024 |
| 131 | 118.5 | 28.0% | — | 1 US$ | 2 US$ | ene 2024 | |
| 132 | Qwen2.5 7B InstructAlibaba Qwen | 118.4 | 35.5% | — | 0,10 US$ | 0,20 US$ | oct 2024 |
| 133 | Ministral 3 3B 2512Mistral AI | 118.1 | 25.3% | — | 0,10 US$ | 0,10 US$ | dic 2025 |
| 134 | 116.5 | 27.0% | — | 0,05 US$ | 0,08 US$ | jul 2024 | |
| 135 | Gemma 3 4BGoogle DeepMind | 116.0 | 23.2% | — | 0,05 US$ | 0,10 US$ | mar 2025 |
| 136 | 102.0 | 23.9% | — | 0,027 US$ | 0,20 US$ | sept 2024 |
| # | Modelo | SWE-bench Verifiedi | Entradai | Salidai | Lanzamiento |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.7Anthropic | 83.5% | 5 US$ | 25 US$ | abr 2026 |
| 2 | Gemini 3.5 FlashGoogle DeepMind | 79.3% | 1,50 US$ | 9 US$ | may 2026 |
| 3 | GLM 5.2Z.ai (Zhipu) | 78.7% | 0,14 US$ | 3,99 US$ | jun 2026 |
| 4 | Claude Opus 4.6Anthropic | 78.7% | 5 US$ | 25 US$ | feb 2026 |
| 5 | DeepSeek V4 Pro 0813DeepSeek | 77.6% | 0,66 US$ | 1,98 US$ | ago 2026 |
| 6 | Qwen3.7 MaxAlibaba Qwen | 77.3% | 1,48 US$ | 4,43 US$ | may 2026 |
| 7 | GPT-5.4OpenAI | 76.9% | 2,50 US$ | 15 US$ | mar 2026 |
| 8 | Qwen3.6 Max PreviewAlibaba Qwen | 76.7% | 1,03 US$ | 6,16 US$ | abr 2026 |
| 9 | Kimi K2.6Moonshot AI | 76.7% | 0,65 US$ | 3,41 US$ | abr 2026 |
| 10 | Claude Opus 4.5Anthropic | 76.7% | 5 US$ | 25 US$ | nov 2025 |
| 11 | Gemini 3.1 Pro Preview Custom ToolsGoogle DeepMind | 75.6% | 2 US$ | 12 US$ | feb 2026 |
| 12 | Gemini 3 Flash PreviewGoogle DeepMind | 75.4% | 0,50 US$ | 3 US$ | dic 2025 |
| 13 | Claude Sonnet 4.6Anthropic | 75.2% | 3 US$ | 15 US$ | feb 2026 |
| 14 | GPT-5.3-CodexOpenAI | 74.8% | 1,75 US$ | 14 US$ | feb 2026 |
| 15 | GLM 5.1Z.ai (Zhipu) | 74.2% | 1,40 US$ | 4,40 US$ | abr 2026 |
| 16 | Kimi K2.5Moonshot AI | 73.8% | 0,45 US$ | 2,25 US$ | ene 2026 |
| 17 | GPT-5.2 ChatOpenAI | 73.8% | 1,75 US$ | 14 US$ | dic 2025 |
| 18 | GPT-5OpenAI | 73.6% | 1,25 US$ | 10 US$ | ago 2025 |
| 19 | Claude Opus 4.1Anthropic | 73.3% | 15 US$ | 75 US$ | ago 2025 |
| 20 | GLM 5Z.ai (Zhipu) | 72.1% | 0,60 US$ | 1,92 US$ | feb 2026 |
| 21 | Claude Sonnet 4.5Anthropic | 71.3% | 3 US$ | 15 US$ | sept 2025 |
| 22 | GPT-5.1OpenAI | 68.0% | 1,25 US$ | 10 US$ | nov 2025 |
| 23 | GPT-5 MiniOpenAI | 64.7% | 0,25 US$ | 2 US$ | ago 2025 |
| 24 | o3OpenAI | 62.3% | 2 US$ | 8 US$ | abr 2025 |
| 25 | Qwen3.6 PlusAlibaba Qwen | 57.9% | 0,33 US$ | 1,95 US$ | abr 2026 |
| 26 | Gemini 2.5 ProGoogle DeepMind | 57.6% | 1,25 US$ | 10 US$ | jun 2025 |
| 27 | GPT-4.1OpenAI | 48.5% | 2 US$ | 8 US$ | abr 2025 |
| 28 | GPT-4o (2024-11-20)OpenAI | 31.0% | 2,50 US$ | 10 US$ | nov 2024 |
| # | Modelo | Matemáticas AIMEi | FrontierMathi | Entradai | Salidai | Lanzamiento |
|---|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5Anthropic | 100% | 88.8% | 2 US$ | 10 US$ | sept 2026 |
| 2 | Claude Opus 5.5Anthropic | 100% | 91.2% | 4 US$ | 20 US$ | sept 2026 |
| 3 | GPT-6 AstraOpenAI | 100% | 93.7% | 10 US$ | 50 US$ | sept 2026 |
| 4 | Qwen3.8 Max (0902)Alibaba Qwen | 100% | 74.7% | 2 US$ | 6 US$ | sept 2026 |
| 5 | Claude Fable 5.1Anthropic | 100% | 90.2% | 10 US$ | 50 US$ | sept 2026 |
| 6 | GPT-5.6 SolOpenAI | 100% | 89.1% | 2 US$ | 10 US$ | jul 2026 |
| 7 | Claude Fable 5Anthropic | 100% | 87.0% | 10 US$ | 50 US$ | jun 2026 |
| 8 | GPT-5.6 TerraOpenAI | 100% | 86.0% | 2 US$ | 12 US$ | jul 2026 |
| 9 | Muse Spark 1.3Meta | 99.2% | 74.4% | 1,25 US$ | 4,25 US$ | sept 2026 |
| 10 | Grok 4.6xAI | 99.2% | 66.0% | 2 US$ | 6 US$ | ago 2026 |
| 11 | Gemini 3.8 FlashGoogle DeepMind | 98.9% | 68.4% | 0,75 US$ | 3,75 US$ | sept 2026 |
| 12 | Claude Opus 5Anthropic | 98.9% | 85.6% | 5 US$ | 25 US$ | jul 2026 |
| 13 | DeepSeek V4 Pro 0813DeepSeek | 98.6% | 64.6% | 0,66 US$ | 1,98 US$ | ago 2026 |
| 14 | GPT-5.6 LunaOpenAI | 98.3% | 82.1% | 0,20 US$ | 1,20 US$ | jul 2026 |
| 15 | Claude Opus 4.8Anthropic | 98.3% | 80.0% | 5 US$ | 25 US$ | may 2026 |
| 16 | Grok 4.5xAI | 97.8% | 57.2% | 2 US$ | 6 US$ | jul 2026 |
| 17 | Claude Opus 4.7Anthropic | 97.8% | 70.2% | 5 US$ | 25 US$ | abr 2026 |
| 18 | GPT-5.4OpenAI | 97.8% | 78.6% | 2,50 US$ | 15 US$ | mar 2026 |
| 19 | Gemini 3.7 FlashGoogle DeepMind | 97.2% | 71.6% | 0,75 US$ | 3,75 US$ | ago 2026 |
| 20 | Kimi K3Moonshot AI | 97.2% | 72.2% | 3 US$ | 15 US$ | jul 2026 |
| 21 | Kimi K2.6Moonshot AI | 96.1% | 57.2% | 0,65 US$ | 3,41 US$ | abr 2026 |
| 22 | GPT-5.2 ChatOpenAI | 96.1% | 67.4% | 1,75 US$ | 14 US$ | dic 2025 |
| 23 | Kimi K2.7 CodeMoonshot AI | 95.6% | 54.0% | 0,66 US$ | 3,30 US$ | jun 2026 |
| 24 | Qwen3.7 MaxAlibaba Qwen | 95.6% | 64.6% | 1,48 US$ | 4,43 US$ | may 2026 |
| 25 | Gemini 3.5 FlashGoogle DeepMind | 95.6% | 62.8% | 1,50 US$ | 9 US$ | may 2026 |
| 26 | Gemini 3.1 Pro PreviewGoogle DeepMind | 95.6% | 59.6% | 2 US$ | 12 US$ | feb 2026 |
| 27 | Gemini 3 Flash PreviewGoogle DeepMind | 95.6% | 51.2% | 0,50 US$ | 3 US$ | dic 2025 |
| 28 | Claude Sonnet 5Anthropic | 94.7% | 65.6% | 2 US$ | 10 US$ | jun 2026 |
| 29 | DeepSeek V4 Flash 0731DeepSeek | 94.4% | 57.5% | 0,018 US$ | 0,32 US$ | jul 2026 |
| 30 | Claude Opus 4.6Anthropic | 94.4% | 66.0% | 5 US$ | 25 US$ | feb 2026 |
| 31 | Gemini 3.6 FlashGoogle DeepMind | 94.2% | 58.9% | 0,75 US$ | 3,75 US$ | jul 2026 |
| 32 | GLM 5.3 FlashZ.ai (Zhipu) | 93.9% | 55.8% | 0,15 US$ | 0,50 US$ | ago 2026 |
| 33 | Qwen3.7 PlusAlibaba Qwen | 93.3% | 34.4% | 0,32 US$ | 1,28 US$ | jun 2026 |
| 34 | Grok 4.3xAI | 93.3% | 42.8% | 1,25 US$ | 2,50 US$ | abr 2026 |
| 35 | GLM 5.1Z.ai (Zhipu) | 93.3% | 36.8% | 1,40 US$ | 4,40 US$ | abr 2026 |
| 36 | Qwen3.6 PlusAlibaba Qwen | 93.3% | 38.2% | 0,33 US$ | 1,95 US$ | abr 2026 |
| 37 | GPT-5OpenAI | 91.4% | 55.4% | 1,25 US$ | 10 US$ | ago 2025 |
| 38 | GLM 5.3Z.ai (Zhipu) | 91.1% | 68.8% | 1,40 US$ | 4,40 US$ | ago 2026 |
| 39 | Qwen3.6 Max PreviewAlibaba Qwen | 91.1% | — | 1,03 US$ | 6,16 US$ | abr 2026 |
| 40 | Qwen3.6 27BAlibaba Qwen | 91.1% | 35.1% | 0,32 US$ | 3,20 US$ | abr 2026 |
| 41 | Inkling SmallThinking Machines Lab | 90.0% | 46.3% | 0,45 US$ | 1,20 US$ | jul 2026 |
| 42 | InklingThinking Machines Lab | 88.9% | 33.3% | 1 US$ | 4,05 US$ | jul 2026 |
| 43 | GPT-5.4 MiniOpenAI | 88.9% | 51.2% | 0,75 US$ | 4,50 US$ | mar 2026 |
| 44 | Qwen3.5 397B A17BAlibaba Qwen | 88.9% | 31.2% | 0,55 US$ | 3,50 US$ | feb 2026 |
| 45 | gpt-oss-120bOpenAI | 88.9% | — | 0,037 US$ | 0,17 US$ | ago 2025 |
| 46 | GPT-5.1OpenAI | 88.6% | — | 1,25 US$ | 10 US$ | nov 2025 |
| 47 | GPT-5.4 NanoOpenAI | 87.8% | 44.9% | 0,20 US$ | 1,25 US$ | mar 2026 |
| 48 | Qwen3.7 FlashAlibaba Qwen | 86.7% | 19.3% | 0,03 US$ | 0,13 US$ | jul 2026 |
| 49 | Nemotron 3 UltraNVIDIA | 86.7% | — | 0,60 US$ | 2,40 US$ | jun 2026 |
| 50 | Qwen3.5 Plus 2026-04-20Alibaba Qwen | 86.7% | — | 0,30 US$ | 1,80 US$ | abr 2026 |
| 51 | Qwen3.6 35B A3BAlibaba Qwen | 86.7% | 20.4% | 0,15 US$ | 1 US$ | abr 2026 |
| 52 | GPT-5 MiniOpenAI | 86.7% | 46.7% | 0,25 US$ | 2 US$ | ago 2025 |
| 53 | Qwen3 235B A22B Thinking 2507Alibaba Qwen | 86.7% | — | 0,23 US$ | 2,30 US$ | jul 2025 |
| 54 | GLM 5.2Z.ai (Zhipu) | 86.4% | 59.2% | 0,14 US$ | 3,99 US$ | jun 2026 |
| 55 | Claude Opus 4.5Anthropic | 86.1% | 34.4% | 5 US$ | 25 US$ | nov 2025 |
| 56 | Claude Sonnet 4.6Anthropic | 85.8% | — | 3 US$ | 15 US$ | feb 2026 |
| 57 | Gemini 2.5 ProGoogle DeepMind | 84.7% | 24.6% | 1,25 US$ | 10 US$ | jun 2025 |
| 58 | Qwen3.6 FlashAlibaba Qwen | 84.4% | 22.5% | 0,19 US$ | 1,13 US$ | abr 2026 |
| 59 | GPT-5.5OpenAI | 84.4% | 85.3% | 5 US$ | 30 US$ | abr 2026 |
| 60 | Qwen3.5-FlashAlibaba Qwen | 84.4% | 18.2% | 0,065 US$ | 0,26 US$ | feb 2026 |
| 61 | o3OpenAI | 84.4% | 33.3% | 2 US$ | 8 US$ | abr 2025 |
| 62 | GLM 4.7Z.ai (Zhipu) | 83.3% | — | 0,60 US$ | 2,20 US$ | dic 2025 |
| 63 | Gemma 4 26B A4B Google DeepMind | 82.2% | — | 0,077 US$ | 0,26 US$ | abr 2026 |
| 64 | o4 MiniOpenAI | 81.7% | 36.1% | 1,10 US$ | 4,40 US$ | abr 2025 |
| 65 | GPT-5 NanoOpenAI | 81.1% | 20.0% | 0,05 US$ | 0,40 US$ | ago 2025 |
| 66 | Gemini 3.1 Flash LiteGoogle DeepMind | 80.0% | 27.7% | 0,25 US$ | 1,50 US$ | may 2026 |
| 67 | GLM 5Z.ai (Zhipu) | 80.0% | — | 0,60 US$ | 1,92 US$ | feb 2026 |
| 68 | Claude Sonnet 4.5Anthropic | 77.8% | 23.9% | 3 US$ | 15 US$ | sept 2025 |
| 69 | o3 MiniOpenAI | 76.9% | 18.6% | 1,10 US$ | 4,40 US$ | ene 2025 |
| 70 | Gemma 4 31BGoogle DeepMind | 73.3% | — | 0,09 US$ | 0,34 US$ | abr 2026 |
| 71 | Qwen3 MaxAlibaba Qwen | 73.3% | 18.9% | 0,78 US$ | 3,90 US$ | sept 2025 |
| 72 | o1OpenAI | 73.3% | 14.7% | 15 US$ | 60 US$ | dic 2024 |
| 73 | Gemini 3.5 Flash LiteGoogle DeepMind | 71.1% | 26.0% | 0,30 US$ | 2,50 US$ | jul 2026 |
| 74 | MiniMax M3MiniMax | 71.1% | — | 0,30 US$ | 1,20 US$ | may 2026 |
| 75 | Claude Sonnet 4Anthropic | 71.1% | — | 3 US$ | 15 US$ | may 2025 |
| 76 | Qwen3 30B A3B Thinking 2507Alibaba Qwen | 70.3% | — | 0,20 US$ | 2,40 US$ | ago 2025 |
| 77 | Qwen3.5-35B-A3BAlibaba Qwen | 70.0% | — | 0,16 US$ | 1,30 US$ | feb 2026 |
| 78 | Claude Opus 4.1Anthropic | 68.9% | 12.6% | 15 US$ | 75 US$ | ago 2025 |
| 79 | Qwen3 32BAlibaba Qwen | 66.9% | — | 0,08 US$ | 0,28 US$ | abr 2025 |
| 80 | Claude Haiku 4.5Anthropic | 66.7% | — | 1 US$ | 5 US$ | oct 2025 |
| 81 | R1 0528DeepSeek | 66.4% | — | 0,50 US$ | 2,15 US$ | may 2025 |
| 82 | Qwen3 14BAlibaba Qwen | 66.4% | — | 0,12 US$ | 0,24 US$ | abr 2025 |
| 83 | gpt-oss-20bOpenAI | 65.3% | — | 0,018 US$ | 0,09 US$ | ago 2025 |
| 84 | Qwen3 30B A3B Instruct 2507Alibaba Qwen | 62.8% | — | 0,048 US$ | 0,19 US$ | jul 2025 |
| 85 | Qwen3.5-9BAlibaba Qwen | 61.7% | — | 0,10 US$ | 0,15 US$ | mar 2026 |
| 86 | GLM 4.7 FlashZ.ai (Zhipu) | 58.3% | — | 0,061 US$ | 0,40 US$ | ene 2026 |
| 87 | Qwen3 8BAlibaba Qwen | 56.1% | — | 0,12 US$ | 0,46 US$ | abr 2025 |
| 88 | DeepSeek V3DeepSeek | 48.9% | — | 0,26 US$ | 1,03 US$ | dic 2024 |
| 89 | GPT-4.1 MiniOpenAI | 44.7% | 6.7% | 0,40 US$ | 1,60 US$ | abr 2025 |
| 90 | GPT-4.1OpenAI | 38.3% | 6.0% | 2 US$ | 8 US$ | abr 2025 |
| 91 | GPT-4.1 NanoOpenAI | 28.9% | — | 0,10 US$ | 0,40 US$ | abr 2025 |
| 92 | Gemma 3 27BGoogle DeepMind | 22.5% | — | 0,08 US$ | 0,45 US$ | mar 2025 |
| 93 | Qwen Plus 0728Alibaba Qwen | 17.8% | — | 0,26 US$ | 0,78 US$ | sept 2025 |
| 94 | Gemma 3 12BGoogle DeepMind | 16.7% | — | 0,05 US$ | 0,15 US$ | mar 2025 |
| 95 | Phi 4Microsoft | 13.8% | — | 0,07 US$ | 0,14 US$ | ene 2025 |
| 96 | Mistral Large 3 2512Mistral AI | 8.5% | — | 0,50 US$ | 1,50 US$ | dic 2025 |
| 97 | Qwen2.5 72B InstructAlibaba Qwen | 8.1% | — | 0,36 US$ | 0,40 US$ | sept 2024 |
| 98 | Gemma 3 4BGoogle DeepMind | 7.5% | — | 0,05 US$ | 0,10 US$ | mar 2025 |
| 99 | GPT-4o-miniOpenAI | 6.9% | 0.7% | 0,15 US$ | 0,60 US$ | jul 2024 |
| 100 | Mistral Small 3Mistral AI | 6.7% | — | 0,05 US$ | 0,08 US$ | ene 2025 |
| 101 | GPT-4 TurboOpenAI | 6.7% | 0.7% | 10 US$ | 30 US$ | abr 2024 |
| 102 | GPT-4o (2024-11-20)OpenAI | 6.4% | 0.4% | 2,50 US$ | 10 US$ | nov 2024 |
| 103 | Mistral Small 4Mistral AI | 5.8% | — | 0,15 US$ | 0,60 US$ | mar 2026 |
| 104 | 5.1% | — | 0,10 US$ | 0,32 US$ | dic 2024 | |
| 105 | 3.6% | — | 0,40 US$ | 0,40 US$ | jul 2024 | |
| 106 | 2.8% | — | 0,017 US$ | 0,11 US$ | oct 2025 | |
| 107 | Qwen2.5 7B InstructAlibaba Qwen | 2.5% | — | 0,10 US$ | 0,20 US$ | oct 2024 |
| 108 | 2.2% | 0.0% | 1 US$ | 2 US$ | ene 2024 | |
| 109 | 1.7% | — | 0,05 US$ | 0,08 US$ | jul 2024 | |
| 110 | Gemma 2 27BGoogle DeepMind | 1.4% | — | 0,65 US$ | 0,65 US$ | jul 2024 |
| 111 | GPT-4OpenAI | 1.1% | — | 30 US$ | 60 US$ | may 2023 |
| 112 | 0.6% | — | 0,027 US$ | 0,20 US$ | sept 2024 |
| # | Modelo | GPQA Diamondi | Entradai | Salidai | Lanzamiento |
|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 95.8% | 10 US$ | 50 US$ | sept 2026 |
| 2 | Claude Sonnet 5.5Anthropic | 95.6% | 2 US$ | 10 US$ | sept 2026 |
| 3 | Gemini 3.8 FlashGoogle DeepMind | 95.4% | 0,75 US$ | 3,75 US$ | sept 2026 |
| 4 | Gemini 3.7 FlashGoogle DeepMind | 94.8% | 0,75 US$ | 3,75 US$ | ago 2026 |
| 5 | GPT-5.4 ProOpenAI | 94.6% | 30 US$ | 180 US$ | mar 2026 |
| 6 | Gemini 3.1 Pro PreviewGoogle DeepMind | 94.4% | 2 US$ | 12 US$ | feb 2026 |
| 7 | Gemini 3.6 FlashGoogle DeepMind | 94.1% | 0,75 US$ | 3,75 US$ | jul 2026 |
| 8 | Grok 4.6xAI | 94.0% | 2 US$ | 6 US$ | ago 2026 |
| 9 | Claude Opus 5Anthropic | 93.9% | 5 US$ | 25 US$ | jul 2026 |
| 10 | GPT-5.6 SolOpenAI | 93.5% | 2 US$ | 10 US$ | jul 2026 |
| 11 | Grok 4.5xAI | 93.4% | 2 US$ | 6 US$ | jul 2026 |
| 12 | GPT-5.6 TerraOpenAI | 93.3% | 2 US$ | 12 US$ | jul 2026 |
| 13 | GPT-5.4OpenAI | 93.3% | 2,50 US$ | 15 US$ | mar 2026 |
| 14 | Kimi K3Moonshot AI | 93.1% | 3 US$ | 15 US$ | jul 2026 |
| 15 | Gemini 3.5 FlashGoogle DeepMind | 92.8% | 1,50 US$ | 9 US$ | may 2026 |
| 16 | Qwen3.8 Max (0902)Alibaba Qwen | 92.7% | 2 US$ | 6 US$ | sept 2026 |
| 17 | GLM 5.2Z.ai (Zhipu) | 91.9% | 0,14 US$ | 3,99 US$ | jun 2026 |
| 18 | DeepSeek V4 Pro 0813DeepSeek | 91.7% | 0,66 US$ | 1,98 US$ | ago 2026 |
| 19 | GPT-5.6 LunaOpenAI | 91.6% | 0,20 US$ | 1,20 US$ | jul 2026 |
| 20 | GPT-5.2 ChatOpenAI | 91.4% | 1,75 US$ | 14 US$ | dic 2025 |
| 21 | DeepSeek V4 Flash 0731DeepSeek | 91.0% | 0,018 US$ | 0,32 US$ | jul 2026 |
| 22 | Claude Opus 4.8Anthropic | 91.0% | 5 US$ | 25 US$ | may 2026 |
| 23 | GLM 5.3Z.ai (Zhipu) | 90.9% | 1,40 US$ | 4,40 US$ | ago 2026 |
| 24 | MiniMax M3MiniMax | 90.9% | 0,30 US$ | 1,20 US$ | may 2026 |
| 25 | Qwen3.7 MaxAlibaba Qwen | 90.9% | 1,48 US$ | 4,43 US$ | may 2026 |
| 26 | Kimi K2.6Moonshot AI | 90.8% | 0,65 US$ | 3,41 US$ | abr 2026 |
| 27 | GPT-5.5OpenAI | 90.7% | 5 US$ | 30 US$ | abr 2026 |
| 28 | Claude Opus 5.5Anthropic | 90.6% | 4 US$ | 20 US$ | sept 2026 |
| 29 | Claude Sonnet 5Anthropic | 90.5% | 2 US$ | 10 US$ | jun 2026 |
| 30 | Claude Opus 4.6Anthropic | 90.5% | 5 US$ | 25 US$ | feb 2026 |
| 31 | GLM 5.3 FlashZ.ai (Zhipu) | 90.2% | 0,15 US$ | 0,50 US$ | ago 2026 |
| 32 | Claude Opus 4.7Anthropic | 90.2% | 5 US$ | 25 US$ | abr 2026 |
| 33 | GLM 5.1Z.ai (Zhipu) | 89.9% | 1,40 US$ | 4,40 US$ | abr 2026 |
| 34 | Gemini 3 Flash PreviewGoogle DeepMind | 89.4% | 0,50 US$ | 3 US$ | dic 2025 |
| 35 | Grok 4.3xAI | 88.8% | 1,25 US$ | 2,50 US$ | abr 2026 |
| 36 | Inkling SmallThinking Machines Lab | 88.5% | 0,45 US$ | 1,20 US$ | jul 2026 |
| 37 | Qwen3.6 PlusAlibaba Qwen | 88.4% | 0,33 US$ | 1,95 US$ | abr 2026 |
| 38 | InklingThinking Machines Lab | 88.3% | 1 US$ | 4,05 US$ | jul 2026 |
| 39 | Kimi K2.7 CodeMoonshot AI | 87.9% | 0,66 US$ | 3,30 US$ | jun 2026 |
| 40 | Qwen3.7 PlusAlibaba Qwen | 87.9% | 0,32 US$ | 1,28 US$ | jun 2026 |
| 41 | GLM 5Z.ai (Zhipu) | 87.8% | 0,60 US$ | 1,92 US$ | feb 2026 |
| 42 | GPT-5.1OpenAI | 87.6% | 1,25 US$ | 10 US$ | nov 2025 |
| 43 | Qwen3.6 Max PreviewAlibaba Qwen | 87.4% | 1,03 US$ | 6,16 US$ | abr 2026 |
| 44 | Claude Sonnet 4.6Anthropic | 87.4% | 3 US$ | 15 US$ | feb 2026 |
| 45 | GPT-5.4 MiniOpenAI | 86.9% | 0,75 US$ | 4,50 US$ | mar 2026 |
| 46 | Qwen3.5 397B A17BAlibaba Qwen | 86.4% | 0,55 US$ | 3,50 US$ | feb 2026 |
| 47 | GPT-5OpenAI | 86.2% | 1,25 US$ | 10 US$ | ago 2025 |
| 48 | Claude Opus 4.5Anthropic | 86.0% | 5 US$ | 25 US$ | nov 2025 |
| 49 | Claude Fable 5Anthropic | 85.9% | 10 US$ | 50 US$ | jun 2026 |
| 50 | Qwen3.6 27BAlibaba Qwen | 85.9% | 0,32 US$ | 3,20 US$ | abr 2026 |
| 51 | Nemotron 3 UltraNVIDIA | 85.4% | 0,60 US$ | 2,40 US$ | jun 2026 |
| 52 | Gemini 2.5 ProGoogle DeepMind | 85.3% | 1,25 US$ | 10 US$ | jun 2025 |
| 53 | Qwen3.5 Plus 2026-04-20Alibaba Qwen | 84.8% | 0,30 US$ | 1,80 US$ | abr 2026 |
| 54 | Qwen3.6 35B A3BAlibaba Qwen | 84.8% | 0,15 US$ | 1 US$ | abr 2026 |
| 55 | Gemini 2.5 Pro Preview 06-05Google DeepMind | 84.8% | 1,25 US$ | 10 US$ | jun 2025 |
| 56 | Qwen3.5-35B-A3BAlibaba Qwen | 83.5% | 0,16 US$ | 1,30 US$ | feb 2026 |
| 57 | Gemini 3.5 Flash LiteGoogle DeepMind | 83.3% | 0,30 US$ | 2,50 US$ | jul 2026 |
| 58 | Qwen3.6 FlashAlibaba Qwen | 83.3% | 0,19 US$ | 1,13 US$ | abr 2026 |
| 59 | GLM 4.7Z.ai (Zhipu) | 83.3% | 0,60 US$ | 2,20 US$ | dic 2025 |
| 60 | Qwen3.7 FlashAlibaba Qwen | 82.3% | 0,03 US$ | 0,13 US$ | jul 2026 |
| 61 | Qwen3.5-FlashAlibaba Qwen | 82.3% | 0,065 US$ | 0,26 US$ | feb 2026 |
| 62 | Claude Sonnet 4.5Anthropic | 82.3% | 3 US$ | 15 US$ | sept 2025 |
| 63 | Gemini 3.1 Flash LiteGoogle DeepMind | 81.8% | 0,25 US$ | 1,50 US$ | may 2026 |
| 64 | o3OpenAI | 81.8% | 2 US$ | 8 US$ | abr 2025 |
| 65 | Qwen3 235B A22B Thinking 2507Alibaba Qwen | 80.1% | 0,23 US$ | 2,30 US$ | jul 2025 |
| 66 | o4 MiniOpenAI | 79.6% | 1,10 US$ | 4,40 US$ | abr 2025 |
| 67 | Claude Sonnet 4Anthropic | 79.2% | 3 US$ | 15 US$ | may 2025 |
| 68 | Qwen3.5-9BAlibaba Qwen | 79.0% | 0,10 US$ | 0,15 US$ | mar 2026 |
| 69 | GPT-5.4 NanoOpenAI | 78.5% | 0,20 US$ | 1,25 US$ | mar 2026 |
| 70 | Claude Opus 4.1Anthropic | 77.3% | 15 US$ | 75 US$ | ago 2025 |
| 71 | o3 MiniOpenAI | 77.0% | 1,10 US$ | 4,40 US$ | ene 2025 |
| 72 | o1OpenAI | 76.8% | 15 US$ | 60 US$ | dic 2024 |
| 73 | R1 0528DeepSeek | 76.3% | 0,50 US$ | 2,15 US$ | may 2025 |
| 74 | Gemma 4 31BGoogle DeepMind | 75.8% | 0,09 US$ | 0,34 US$ | abr 2026 |
| 75 | gpt-oss-120bOpenAI | 75.8% | 0,037 US$ | 0,17 US$ | ago 2025 |
| 76 | GPT-5 MiniOpenAI | 75.0% | 0,25 US$ | 2 US$ | ago 2025 |
| 77 | Gemma 4 26B A4B Google DeepMind | 73.2% | 0,077 US$ | 0,26 US$ | abr 2026 |
| 78 | Qwen3 MaxAlibaba Qwen | 72.6% | 0,78 US$ | 3,90 US$ | sept 2025 |
| 79 | Claude Haiku 4.5Anthropic | 71.2% | 1 US$ | 5 US$ | oct 2025 |
| 80 | DeepSeek V3DeepSeek | 71.2% | 0,26 US$ | 1,03 US$ | dic 2024 |
| 81 | Qwen3 235B A22B Instruct 2507Alibaba Qwen | 70.7% | 0,088 US$ | 0,35 US$ | jul 2025 |
| 82 | Qwen3 30B A3B Thinking 2507Alibaba Qwen | 70.1% | 0,20 US$ | 2,40 US$ | ago 2025 |
| 83 | GPT-5 NanoOpenAI | 69.4% | 0,05 US$ | 0,40 US$ | ago 2025 |
| 84 | GPT-4.1OpenAI | 66.9% | 2 US$ | 8 US$ | abr 2025 |
| 85 | GPT-4.1 MiniOpenAI | 65.8% | 0,40 US$ | 1,60 US$ | abr 2025 |
| 86 | Qwen3 32BAlibaba Qwen | 65.7% | 0,08 US$ | 0,28 US$ | abr 2025 |
| 87 | Qwen3 14BAlibaba Qwen | 63.8% | 0,12 US$ | 0,24 US$ | abr 2025 |
| 88 | Qwen3 30B A3B Instruct 2507Alibaba Qwen | 61.7% | 0,048 US$ | 0,19 US$ | jul 2025 |
| 89 | gpt-oss-20bOpenAI | 60.8% | 0,018 US$ | 0,09 US$ | ago 2025 |
| 90 | GLM 4.7 FlashZ.ai (Zhipu) | 60.5% | 0,061 US$ | 0,40 US$ | ene 2026 |
| 91 | Qwen3 8BAlibaba Qwen | 56.8% | 0,12 US$ | 0,46 US$ | abr 2025 |
| 92 | Phi 4Microsoft | 56.1% | 0,07 US$ | 0,14 US$ | ene 2025 |
| 93 | Mistral Large 3 2512Mistral AI | 51.3% | 0,50 US$ | 1,50 US$ | dic 2025 |
| 94 | GPT-4o (2024-11-20)OpenAI | 49.2% | 2,50 US$ | 10 US$ | nov 2024 |
| 95 | Qwen2.5 72B InstructAlibaba Qwen | 49.1% | 0,36 US$ | 0,40 US$ | sept 2024 |
| 96 | GPT-4.1 NanoOpenAI | 48.9% | 0,10 US$ | 0,40 US$ | abr 2025 |
| 97 | Qwen Plus 0728Alibaba Qwen | 48.1% | 0,26 US$ | 0,78 US$ | sept 2025 |
| 98 | Gemma 3 27BGoogle DeepMind | 47.7% | 0,08 US$ | 0,45 US$ | mar 2025 |
| 99 | Mistral Small 4Mistral AI | 47.5% | 0,15 US$ | 0,60 US$ | mar 2026 |
| 100 | 47.4% | 0,10 US$ | 0,32 US$ | dic 2024 | |
| 101 | Mistral Small 3Mistral AI | 47.3% | 0,05 US$ | 0,08 US$ | ene 2025 |
| 102 | GPT-4 TurboOpenAI | 46.6% | 10 US$ | 30 US$ | abr 2024 |
| 103 | 44.2% | 0,40 US$ | 0,40 US$ | jul 2024 | |
| 104 | WizardLM-2 8x22BMicrosoft | 43.4% | 0,62 US$ | 0,62 US$ | abr 2024 |
| 105 | Gemma 3 12BGoogle DeepMind | 39.5% | 0,05 US$ | 0,15 US$ | mar 2025 |
| 106 | GPT-4o-miniOpenAI | 37.7% | 0,15 US$ | 0,60 US$ | jul 2024 |
| 107 | Gemma 2 27BGoogle DeepMind | 36.5% | 0,65 US$ | 0,65 US$ | jul 2024 |
| 108 | GPT-4OpenAI | 35.7% | 30 US$ | 60 US$ | may 2023 |
| 109 | Qwen2.5 7B InstructAlibaba Qwen | 35.5% | 0,10 US$ | 0,20 US$ | oct 2024 |
| 110 | 28.3% | 0,017 US$ | 0,11 US$ | oct 2025 | |
| 111 | 28.0% | 1 US$ | 2 US$ | ene 2024 | |
| 112 | Ministral 3 8B 2512Mistral AI | 27.1% | 0,15 US$ | 0,15 US$ | dic 2025 |
| 113 | 27.0% | 0,05 US$ | 0,08 US$ | jul 2024 | |
| 114 | Ministral 3 3B 2512Mistral AI | 25.3% | 0,10 US$ | 0,10 US$ | dic 2025 |
| 115 | 23.9% | 0,027 US$ | 0,20 US$ | sept 2024 | |
| 116 | Gemma 3 4BGoogle DeepMind | 23.2% | 0,05 US$ | 0,10 US$ | mar 2025 |
| # | Modelo | SimpleQA Verifiedi | Entradai | Salidai | Lanzamiento |
|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 75.6% | 10 US$ | 50 US$ | sept 2026 |
| 2 | Gemini 3.1 Pro PreviewGoogle DeepMind | 73.5% | 2 US$ | 12 US$ | feb 2026 |
| 3 | Claude Opus 5.5Anthropic | 72.2% | 4 US$ | 20 US$ | sept 2026 |
| 4 | Claude Fable 5.1Anthropic | 70.8% | 10 US$ | 50 US$ | sept 2026 |
| 5 | Claude Fable 5Anthropic | 70.7% | 10 US$ | 50 US$ | jun 2026 |
| 6 | Gemini 3.8 FlashGoogle DeepMind | 69.7% | 0,75 US$ | 3,75 US$ | sept 2026 |
| 7 | GPT-5.6 SolOpenAI | 69.7% | 2 US$ | 10 US$ | jul 2026 |
| 8 | Gemini 3.7 FlashGoogle DeepMind | 69.2% | 0,75 US$ | 3,75 US$ | ago 2026 |
| 9 | Gemini 3 Flash PreviewGoogle DeepMind | 66.8% | 0,50 US$ | 3 US$ | dic 2025 |
| 10 | Gemini 3.6 FlashGoogle DeepMind | 66.2% | 0,75 US$ | 3,75 US$ | jul 2026 |
| 11 | Gemini 3.5 FlashGoogle DeepMind | 66.2% | 1,50 US$ | 9 US$ | may 2026 |
| 12 | GPT-5.5OpenAI | 63.0% | 5 US$ | 30 US$ | abr 2026 |
| 13 | Muse Spark 1.2Meta | 60.3% | 1,25 US$ | 4,25 US$ | ago 2026 |
| 14 | Claude Opus 5Anthropic | 59.9% | 5 US$ | 25 US$ | jul 2026 |
| 15 | Muse Spark 1.1Meta | 57.8% | 1,25 US$ | 4,25 US$ | jul 2026 |
| 16 | Qwen3.7 MaxAlibaba Qwen | 55.8% | 1,48 US$ | 4,43 US$ | may 2026 |
| 17 | Claude Opus 4.8Anthropic | 53.0% | 5 US$ | 25 US$ | may 2026 |
| 18 | DeepSeek V4 Pro 0813DeepSeek | 52.9% | 0,66 US$ | 1,98 US$ | ago 2026 |
| 19 | Qwen3.6 Max PreviewAlibaba Qwen | 52.0% | 1,03 US$ | 6,16 US$ | abr 2026 |
| 20 | Claude Opus 4.7Anthropic | 51.7% | 5 US$ | 25 US$ | abr 2026 |
| 21 | Kimi K3Moonshot AI | 50.6% | 3 US$ | 15 US$ | jul 2026 |
| 22 | GPT-5OpenAI | 50.1% | 1,25 US$ | 10 US$ | ago 2025 |
| 23 | o3OpenAI | 49.4% | 2 US$ | 8 US$ | abr 2025 |
| 24 | Grok 4.6xAI | 49.3% | 2 US$ | 6 US$ | ago 2026 |
| 25 | Qwen3 MaxAlibaba Qwen | 48.7% | 0,78 US$ | 3,90 US$ | sept 2025 |
| 26 | Grok 4.5xAI | 48.3% | 2 US$ | 6 US$ | jul 2026 |
| 27 | GPT-5.1OpenAI | 48.0% | 1,25 US$ | 10 US$ | nov 2025 |
| 28 | Qwen3.8 Max (0902)Alibaba Qwen | 47.3% | 2 US$ | 6 US$ | sept 2026 |
| 29 | Claude Opus 4.6Anthropic | 47.0% | 5 US$ | 25 US$ | feb 2026 |
| 30 | Claude Sonnet 5.5Anthropic | 46.5% | 2 US$ | 10 US$ | sept 2026 |
| 31 | GPT-5.4 ProOpenAI | 46.3% | 30 US$ | 180 US$ | mar 2026 |
| 32 | Claude Opus 4.5Anthropic | 45.7% | 5 US$ | 25 US$ | nov 2025 |
| 33 | GPT-5.4OpenAI | 45.1% | 2,50 US$ | 15 US$ | mar 2026 |
| 34 | Qwen3.6 PlusAlibaba Qwen | 44.1% | 0,33 US$ | 1,95 US$ | abr 2026 |
| 35 | GPT-5.6 TerraOpenAI | 43.2% | 2 US$ | 12 US$ | jul 2026 |
| 36 | o1OpenAI | 41.1% | 15 US$ | 60 US$ | dic 2024 |
| 37 | GLM 5.3Z.ai (Zhipu) | 41.0% | 1,40 US$ | 4,40 US$ | ago 2026 |
| 38 | GPT-5.6 LunaOpenAI | 41.0% | 0,20 US$ | 1,20 US$ | jul 2026 |
| 39 | Qwen3 235B A22B Thinking 2507Alibaba Qwen | 40.4% | 0,23 US$ | 2,30 US$ | jul 2025 |
| 40 | InklingThinking Machines Lab | 40.3% | 1 US$ | 4,05 US$ | jul 2026 |
| 41 | GPT-5.2 ChatOpenAI | 37.1% | 1,75 US$ | 14 US$ | dic 2025 |
| 42 | Kimi K2.7 CodeMoonshot AI | 36.5% | 0,66 US$ | 3,30 US$ | jun 2026 |
| 43 | Claude Sonnet 4.6Anthropic | 35.5% | 3 US$ | 15 US$ | feb 2026 |
| 44 | Kimi K2.6Moonshot AI | 34.9% | 0,65 US$ | 3,41 US$ | abr 2026 |
| 45 | Kimi K2.5Moonshot AI | 34.3% | 0,45 US$ | 2,25 US$ | ene 2026 |
| 46 | GLM 5.2Z.ai (Zhipu) | 34.2% | 0,14 US$ | 3,99 US$ | jun 2026 |
| 47 | GLM 5.1Z.ai (Zhipu) | 34.0% | 1,40 US$ | 4,40 US$ | abr 2026 |
| 48 | Claude Sonnet 5Anthropic | 33.7% | 2 US$ | 10 US$ | jun 2026 |
| 49 | DeepSeek V4 Flash 0731DeepSeek | 33.6% | 0,018 US$ | 0,32 US$ | jul 2026 |
| 50 | Grok 4.3xAI | 33.2% | 1,25 US$ | 2,50 US$ | abr 2026 |
| 51 | GLM 4.7Z.ai (Zhipu) | 32.2% | 0,60 US$ | 2,20 US$ | dic 2025 |
| 52 | GPT-4.1OpenAI | 31.1% | 2 US$ | 8 US$ | abr 2025 |
| 53 | Claude Sonnet 4.5Anthropic | 30.7% | 3 US$ | 15 US$ | sept 2025 |
| 54 | GPT-5.4 MiniOpenAI | 29.4% | 0,75 US$ | 4,50 US$ | mar 2026 |
| 55 | GPT-4o (2024-11-20)OpenAI | 26.0% | 2,50 US$ | 10 US$ | nov 2024 |
| 56 | Qwen3.5 Plus 2026-04-20Alibaba Qwen | 25.4% | 0,30 US$ | 1,80 US$ | abr 2026 |
| 57 | GPT-5 MiniOpenAI | 21.6% | 0,25 US$ | 2 US$ | ago 2025 |
| 58 | Qwen3.5-FlashAlibaba Qwen | 20.3% | 0,065 US$ | 0,26 US$ | feb 2026 |
| 59 | o4 MiniOpenAI | 19.6% | 1,10 US$ | 4,40 US$ | abr 2025 |
| 60 | Inkling SmallThinking Machines Lab | 19.1% | 0,45 US$ | 1,20 US$ | jul 2026 |
| 61 | Qwen3.6 FlashAlibaba Qwen | 15.9% | 0,19 US$ | 1,13 US$ | abr 2026 |
| 62 | o3 MiniOpenAI | 15.3% | 1,10 US$ | 4,40 US$ | ene 2025 |
| 63 | Claude Haiku 4.5Anthropic | 13.2% | 1 US$ | 5 US$ | oct 2025 |
| 64 | GPT-4.1 MiniOpenAI | 12.7% | 0,40 US$ | 1,60 US$ | abr 2025 |
| 65 | GPT-5.4 NanoOpenAI | 11.7% | 0,20 US$ | 1,25 US$ | mar 2026 |
| 66 | GPT-5 NanoOpenAI | 11.7% | 0,05 US$ | 0,40 US$ | ago 2025 |
| 67 | Gemma 4 31BGoogle DeepMind | 10.4% | 0,09 US$ | 0,34 US$ | abr 2026 |
| 68 | GPT-4o-miniOpenAI | 8.3% | 0,15 US$ | 0,60 US$ | jul 2024 |
| 69 | GPT-4.1 NanoOpenAI | 6.0% | 0,10 US$ | 0,40 US$ | abr 2025 |
Sin resultados. Prueba a quitar un filtro.
por millón de tokens · USD
Cómo leer estos rankings
Las puntuaciones provienen de pruebas realizadas de forma independiente por Epoch AI, no de cifras publicadas por las propias empresas de IA. Cuando un modelo se probó con varias configuraciones (por ejemplo, distintos niveles de razonamiento), mostramos su mejor resultado. Los modelos muy nuevos aparecen cuando ya se han evaluado, normalmente pocas semanas después de su lanzamiento.
- Índice general
- Epoch Capabilities Index: una sola puntuación que combina decenas de benchmarks, para poder comparar modelos evaluados con pruebas distintas. Cuanto más alto, mejor; es un índice, no un porcentaje.
- SWE-bench Verified
- ¿Puede el modelo corregir errores reales en proyectos reales de software de código abierto? Porcentaje de tareas resueltas.
- GPQA Diamond
- Preguntas de opción múltiple muy difíciles de biología, física y química, escritas por expertos con doctorado. Responder al azar da un 25 %.
- Matemáticas AIME
- Problemas de matemáticas de nivel de competición, al estilo del American Invitational Mathematics Examination. Porcentaje resuelto.
- FrontierMath
- Problemas de matemáticas originales e inéditos escritos por matemáticos profesionales, desde nivel avanzado hasta nivel de investigación. Porcentaje resuelto.
- SimpleQA Verified
- Preguntas factuales cortas. Mide con qué frecuencia el modelo acierta los datos en lugar de inventárselos.
Datos de benchmarks: Epoch AI, con licencia CC BY 4.0 · Última actualización: 30 sept 2026 · Cómo trabajamos
ProgramaciónMatemáticasCienciasBaratosGratisCódigo abiertoContexto largoImágenes y PDF