📚 Clasificación: Conocimiento
Conocimiento factual y anti-alucinación: SimpleQA, MMLU-Pro, MMMLU.
🏆 General673💻 Código392📐 Matemáticas32🧠 Razonamiento391🤖 Agéntico31📚 Conocimiento388👁️ Multimodal370🛠️ Agentes de código40🔬 Razonamiento científico45🌍 Multilingüe12✍️ Escritura53💬 Chat y asistentes53🔭 Investigación48🔧 Llamada de herramientas31📜 Contexto largo55⚡ Velocidad378💰 Valor377🔓 Pesos abiertos153
388 modelos
| # | Modelo | Podium Score | Velocidad |
|---|---|---|---|
| 01 | Claude Mythos Preview Anthropic · preview | 100.0 | 80 t/s |
| 02 | Gemini 3.1 Pro Google | 97.6 | 136 t/s |
| 03 | Gemini 3 Pro Google | 92.9 | 120 t/s |
| 4 | Claude Mythos 5 Anthropic · preview | 75.0 | 50 t/s |
| 5 | Qwen3.7 Max Alibaba | 69.3 | 203 t/s |
| 6 | Claude Sonnet 4.6 (max) Anthropic | 67.0 | 50 t/s |
| 7 | GPT-5.5 Pro OpenAI | 50.0 | 60 t/s |
| 8 | Qwen3.6 Plus Alibaba | 48.4 | 55 t/s |
| 9 | Gpt 5 6 Terra High OpenAI | 45.0 | 90 t/s |
| 10 | Claude Sonnet 4 6 Adaptive Anthropic | 45.0 | 80 t/s |
| 11 | Gpt 5 6 Luna High OpenAI | 45.0 | 90 t/s |
| 12 | Gpt 5 6 Terra Medium OpenAI | 45.0 | 90 t/s |
| 13 | Deepseek V4 Pro 0424 DeepSeek · Abierto | 45.0 | 60 t/s |
| 14 | Claude Opus 4 6 Adaptive Anthropic | 45.0 | 80 t/s |
| 15 | Gpt 5 5 Low OpenAI | 45.0 | 90 t/s |
| 16 | Deepseek V4 Pro 0424 High DeepSeek · Abierto | 45.0 | 60 t/s |
| 17 | Claude Opus 4 5 Thinking Anthropic | 45.0 | 50 t/s |
| 18 | Gpt 5 6 Terra Low OpenAI | 45.0 | 90 t/s |
| 19 | Gpt 5 2 Codex OpenAI | 45.0 | 90 t/s |
| 20 | Qwen3 6 Max Alibaba · Abierto | 45.0 | 85 t/s |
| 21 | 45.0 | 80 t/s | |
| 22 | Gpt 5 4 Low OpenAI | 45.0 | 90 t/s |
| 23 | Agnes 2 5 Pro Alpha Agnes | 45.0 | 60 t/s |
| 24 | Gpt 5 2 Medium OpenAI | 45.0 | 90 t/s |
| 25 | Gpt 5 6 Luna Medium OpenAI | 45.0 | 90 t/s |
| 26 | Deepseek V3 2 Reasoning DeepSeek · Abierto | 45.0 | 40 t/s |
| 27 | Grok 4 20 xAI | 45.0 | 80 t/s |
| 28 | Claude 4 5 Sonnet Thinking Anthropic | 45.0 | 50 t/s |
| 29 | 45.0 | 80 t/s | |
| 30 | Gpt 5 Codex OpenAI | 45.0 | 90 t/s |
| 31 | Grok 4 3 Low xAI | 45.0 | 80 t/s |
| 32 | Gpt 5 1 Codex OpenAI | 45.0 | 90 t/s |
| 33 | Gpt 5 Medium OpenAI | 45.0 | 90 t/s |
| 34 | Claude 4 1 Opus Thinking Anthropic | 45.0 | 50 t/s |
| 35 | Hy3 Preview Tencent | 45.0 | 60 t/s |
| 36 | Gpt 5 5 Instant 05 26 OpenAI | 45.0 | 90 t/s |
| 37 | Gpt 5 6 Luna Low OpenAI | 45.0 | 90 t/s |
| 38 | Kimi K2 Thinking Moonshot AI | 45.0 | 40 t/s |
| 39 | O3 Pro OpenAI | 45.0 | 60 t/s |
| 40 | Claude 4 5 Haiku Reasoning Anthropic | 45.0 | 50 t/s |
| 41 | 45.0 | 40 t/s | |
| 42 | Gpt 5 Low OpenAI | 45.0 | 90 t/s |
| 43 | Claude 4 Opus Thinking Anthropic | 45.0 | 50 t/s |
| 44 | Ring 2 6 1t WeRide | 45.0 | 60 t/s |
| 45 | Deepseek V3 1 Terminus Reasoning DeepSeek · Abierto | 45.0 | 40 t/s |
| 46 | Claude 4 Sonnet Thinking Anthropic | 45.0 | 50 t/s |
| 47 | Glm 4 6 Reasoning Z.ai · Abierto | 45.0 | 40 t/s |
| 48 | Gpt 5 5 Instant 06 26 OpenAI | 45.0 | 90 t/s |
| 49 | Claude 4 1 Opus Anthropic | 45.0 | 80 t/s |
| 50 | 45.0 | 50 t/s | |
| 51 | Claude 4 Opus Anthropic | 45.0 | 80 t/s |
| 52 | Deepseek V3 2 Reasoning 0925 DeepSeek · Abierto | 45.0 | 40 t/s |
| 53 | Qwen3 Max Thinking Preview Alibaba · Abierto | 45.0 | 55 t/s |
| 54 | O1 OpenAI | 45.0 | 60 t/s |
| 55 | Gemma 4 12b Google · Abierto | 45.0 | 120 t/s |
| 56 | Mercury 2 Inception | 45.0 | 60 t/s |
| 57 | Qwen3 5 9b Alibaba · Abierto | 45.0 | 85 t/s |
| 58 | Apriel V1 5 15b Thinker ServiceNow | 45.0 | 60 t/s |
| 59 | 45.0 | 40 t/s | |
| 60 | Deepseek V3 1 Reasoning DeepSeek · Abierto | 45.0 | 40 t/s |
| 61 | Qwen3 Vl 235b A22b Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 62 | Apriel V1 6 15b Thinker ServiceNow | 45.0 | 60 t/s |
| 63 | Qwen3 5 9b Non Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 64 | Qwen3 5 4b Alibaba · Abierto | 45.0 | 85 t/s |
| 65 | Qwen3 235b A22b Instruct 2507 Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 66 | 45.0 | 40 t/s | |
| 67 | O1 Pro OpenAI | 45.0 | 60 t/s |
| 68 | Deepseek R1 0120 DeepSeek · Abierto | 45.0 | 40 t/s |
| 69 | Seed Oss 36b Instruct ByteDance | 45.0 | 60 t/s |
| 70 | Trinity Large Thinking Trinity | 45.0 | 40 t/s |
| 71 | Qwen3 Vl 32b Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 72 | O1 Preview OpenAI | 45.0 | 60 t/s |
| 73 | 45.0 | 40 t/s | |
| 74 | Grok 4 Fast xAI | 45.0 | 80 t/s |
| 75 | Qwen3 5 4b Non Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 76 | Gpt Oss 120b Low OpenAI | 45.0 | 90 t/s |
| 77 | Qwen3 30b A3b 2507 Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 78 | Gpt Oss 20b Low OpenAI | 45.0 | 90 t/s |
| 79 | Qwen3 235b A22b Instruct Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 80 | Qwen3 Vl 30b A3b Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 81 | Qwq 32b Alibaba | 45.0 | 85 t/s |
| 82 | Celeris 1 Celeris | 45.0 | 60 t/s |
| 83 | Claude 3 Opus Anthropic | 45.0 | 80 t/s |
| 84 | Nova 2 0 Omni Amazon | 45.0 | 60 t/s |
| 85 | Claude 35 Sonnet Anthropic | 45.0 | 80 t/s |
| 86 | Qwen3 Omni 30b A3b Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 87 | Qwen3 30b A3b Instruct Reasoning Alibaba · Abierto | 45.0 | 55 t/s |
| 88 | Llama 3 1 Nemotron Ultra 253b V1 Reasoning Meta · Abierto | 45.0 | 50 t/s |
| 89 | Hermes 4 Llama 3 1 405b Reasoning Meta · Abierto | 45.0 | 50 t/s |
| 90 | Qwen3 32b Instruct Alibaba · Abierto | 45.0 | 85 t/s |
| 91 | Gpt 4o 2024 05 13 OpenAI | 45.0 | 90 t/s |
| 92 | Claude 35 Sonnet June 24 Anthropic | 45.0 | 80 t/s |
| 93 | Gpt 4 Turbo OpenAI | 45.0 | 90 t/s |
| 94 | Mistral Large 2407 Mistral · Abierto | 45.0 | 100 t/s |
| 95 | Mistral Small 3 Mistral · Abierto | 45.0 | 100 t/s |
| 96 | Qwen Turbo Alibaba · Abierto | 45.0 | 85 t/s |
| 97 | Gemma 3 12b Google · Abierto | 45.0 | 120 t/s |
| 98 | Hermes 3 Llama 3 1 70b Meta · Abierto | 45.0 | 80 t/s |
| 99 | Jamba 1 5 Large AI21 Labs | 45.0 | 60 t/s |
| 100 | Qwen3 Omni 30b A3b Instruct Alibaba · Abierto | 45.0 | 85 t/s |
| 101 | Mistral Small Mistral · Abierto | 45.0 | 100 t/s |
| 102 | Mistral Large Mistral · Abierto | 45.0 | 100 t/s |
| 103 | Gpt 35 Turbo OpenAI | 45.0 | 90 t/s |
| 104 | Mistral Medium Mistral · Abierto | 45.0 | 100 t/s |
| 105 | Mistral Small 2402 Mistral · Abierto | 45.0 | 100 t/s |
| 106 | Command R Plus 04 2024 Cohere | 45.0 | 60 t/s |
| 107 | Mixtral 8x7b Instruct Mistral | 45.0 | 100 t/s |
| 108 | Command R 03 2024 Cohere | 45.0 | 60 t/s |
| 109 | Gemma 3n E2b Google · Abierto | 45.0 | 120 t/s |
| 110 | Gemma 3n E4b Google · Abierto | 45.0 | 120 t/s |
| 111 | Amazon Nova Amazon | 45.0 | 60 t/s |
| 112 | Async Pro V1 0 Async | 45.0 | 60 t/s |
| 113 | Asyncflow V2 Async | 45.0 | 60 t/s |
| 114 | Claude Sonnet 5 High Anthropic | 45.0 | 80 t/s |
| 115 | Claude Sonnet 5 Low Anthropic | 45.0 | 80 t/s |
| 116 | Claude Sonnet 5 Medium Anthropic | 45.0 | 80 t/s |
| 117 | Claude Sonnet 5 Xhigh Anthropic | 45.0 | 80 t/s |
| 118 | Cogito V2 1 Reasoning Cogito | 45.0 | 40 t/s |
| 119 | Falcon 2 TII · Abierto | 45.0 | 60 t/s |
| 120 | Gladia Solaria Upstage | 45.0 | 60 t/s |
| 121 | Gpt 5 4 Pro OpenAI | 45.0 | 90 t/s |
| 122 | Gpt Realtime 2 OpenAI | 45.0 | 90 t/s |
| 123 | Grok Stt xAI | 45.0 | 80 t/s |
| 124 | Hidream O1 Image OpenAI | 45.0 | 60 t/s |
| 125 | Mistral Mistral · Abierto | 45.0 | 100 t/s |
| 126 | Nvidia NVIDIA | 45.0 | 150 t/s |
| 127 | 45.0 | 150 t/s | |
| 128 | Ocateve 2 Ocat | 45.0 | 60 t/s |
| 129 | Qwen Image Alibaba · Abierto | 45.0 | 85 t/s |
| 130 | Qwen3 Asr Alibaba · Abierto | 45.0 | 85 t/s |
| 131 | Claude Fable 5 Anthropic | 43.3 | 71 t/s |
| 132 | ERNIE 5.1 Baidu | 41.7 | 60 t/s |
| 133 | Nemotron 3 Ultra 550B NVIDIA · Abierto | 37.5 | 142 t/s |
| 134 | Claude Opus 5 Anthropic | 37.1 | 60 t/s |
| 135 | Claude Opus 5 Xhigh Anthropic | 35.4 | 80 t/s |
| 136 | Qwen3.5 122B-A10B Alibaba · Abierto | 35.4 | 126 t/s |
| 137 | Solar Pro 3 Upstage | 34.0 | 40 t/s |
| 138 | Claude Opus 5 High Anthropic | 33.7 | 80 t/s |
| 139 | GPT-oss-20B OpenAI · Abierto | 32.0 | 90 t/s |
| 140 | Claude Opus 5 Medium Anthropic | 31.0 | 80 t/s |
| 141 | Grok 4 xAI | 30.5 | 80 t/s |
| 142 | Grok 4 6 xAI | 30.5 | 80 t/s |
| 143 | Gpt 5 4 Mini OpenAI | 30.0 | 90 t/s |
| 144 | Deepseek V4 Flash 0420 DeepSeek · Abierto | 30.0 | 60 t/s |
| 145 | Gpt 5 4 Nano OpenAI | 30.0 | 90 t/s |
| 146 | Deepseek V4 Flash 0420 High DeepSeek · Abierto | 30.0 | 60 t/s |
| 147 | Gemini 3 Flash Reasoning Google | 30.0 | 80 t/s |
| 148 | Gemini 3 Pro Low Google | 30.0 | 120 t/s |
| 149 | Minimax M2 1 MiniMax · Abierto | 30.0 | 60 t/s |
| 150 | Mimo V2 Flash Reasoning Xiaomi · Abierto | 30.0 | 40 t/s |
| 151 | Gpt 5 1 Codex Mini OpenAI | 30.0 | 90 t/s |
| 152 | Step 3 7 Flash StepFun | 30.0 | 60 t/s |
| 153 | Gpt 5 4 Nano Medium OpenAI | 30.0 | 90 t/s |
| 154 | Gpt 5 4 Mini Medium OpenAI | 30.0 | 90 t/s |
| 155 | Minimax M2 MiniMax · Abierto | 30.0 | 60 t/s |
| 156 | O4 Mini OpenAI | 30.0 | 60 t/s |
| 157 | Step 3 5 Flash 0202 StepFun | 30.0 | 60 t/s |
| 158 | Gpt 5 Mini OpenAI | 30.0 | 90 t/s |
| 159 | Gemini 2 5 Pro 05 06 Google | 30.0 | 120 t/s |
| 160 | Nova 2 0 Lite Reasoning Amazon | 30.0 | 40 t/s |
| 161 | 30.0 | 80 t/s | |
| 162 | 30.0 | 40 t/s | |
| 163 | O3 Mini OpenAI | 30.0 | 60 t/s |
| 164 | Qwen3 5 Omni Flash Alibaba · Abierto | 30.0 | 85 t/s |
| 165 | 30.0 | 40 t/s | |
| 166 | Minimax M1 80k MiniMax · Abierto | 30.0 | 60 t/s |
| 167 | O3 Mini High OpenAI | 30.0 | 60 t/s |
| 168 | 30.0 | 80 t/s | |
| 169 | 30.0 | 150 t/s | |
| 170 | Gpt 4 1 Mini OpenAI | 30.0 | 90 t/s |
| 171 | 30.0 | 100 t/s | |
| 172 | Gpt 5 Mini Minimal OpenAI | 30.0 | 90 t/s |
| 173 | Nova 2 0 Lite Amazon | 30.0 | 60 t/s |
| 174 | 30.0 | 80 t/s | |
| 175 | 30.0 | 120 t/s | |
| 176 | Qwen3 Vl 8b Reasoning Alibaba · Abierto | 30.0 | 55 t/s |
| 177 | Gpt 4 1 Nano OpenAI | 30.0 | 90 t/s |
| 178 | 30.0 | 100 t/s | |
| 179 | Qwen3 8b Instruct Reasoning Alibaba · Abierto | 30.0 | 55 t/s |
| 180 | 30.0 | 150 t/s | |
| 181 | Gemini 2 5 Flash Lite Google | 30.0 | 120 t/s |
| 182 | Gpt 4o Mini OpenAI | 30.0 | 90 t/s |
| 183 | Reka Flash Reka | 30.0 | 60 t/s |
| 184 | Solar Mini Upstage | 30.0 | 60 t/s |
| 185 | Qwen3 8b Instruct Alibaba · Abierto | 30.0 | 85 t/s |
| 186 | 30.0 | 150 t/s | |
| 187 | Llama 2 Chat 7b Meta · Abierto | 30.0 | 80 t/s |
| 188 | Reka Flash 3 Reka | 30.0 | 60 t/s |
| 189 | Olmo 3 7b Instruct AI2 · Abierto | 30.0 | 60 t/s |
| 190 | Jamba 1 5 Mini AI21 Labs | 30.0 | 60 t/s |
| 191 | Granite 3 3 8b Instruct IBM · Abierto | 30.0 | 60 t/s |
| 192 | Apertus 8b Instruct Swiss AI | 30.0 | 60 t/s |
| 193 | Llama 3 Instruct 8b Meta · Abierto | 30.0 | 80 t/s |
| 194 | Async Flash V1 5 Async | 30.0 | 60 t/s |
| 195 | Gemini Google | 30.0 | 120 t/s |
| 196 | 30.0 | 120 t/s | |
| 197 | Gemini_nano Banana Pro Google | 30.0 | 120 t/s |
| 198 | Minimax H3 MiniMax · Abierto | 30.0 | 60 t/s |
| 199 | Qwen3.5 397B-A17B Alibaba · Abierto | 29.2 | 71 t/s |
| 200 | Claude Opus 4.8 Anthropic | 28.8 | 50 t/s |
| 201 | Claude Opus 5 Low Anthropic | 28.6 | 80 t/s |
| 202 | Muse Spark 1.1 Meta | 28.1 | 208 t/s |
| 203 | Claude Opus 4.7 Anthropic | 27.3 | 80 t/s |
| 204 | Muse Spark 1 2 Meta | 27.2 | 80 t/s |
| 205 | Muse Spark Meta | 27.2 | 80 t/s |
| 206 | Gemini 3 7 Flash Google | 26.5 | 120 t/s |
| 207 | GPT-5.2 OpenAI | 26.2 | 90 t/s |
| 208 | Grok 4.5 xAI | 25.3 | 59 t/s |
| 209 | Gemini 3 7 Flash Medium Google | 23.7 | 120 t/s |
| 210 | Gemini 3.6 Flash Google | 22.1 | 233 t/s |
| 211 | Gemini 3 7 Flash Low Google | 22.1 | 120 t/s |
| 212 | GPT-5.6 Sol OpenAI | 22.0 | 72 t/s |
| 213 | GPT-5 OpenAI | 22.0 | 60 t/s |
| 214 | Gemini 3.5 Flash Google | 21.2 | 267 t/s |
| 215 | GPT-5.5 OpenAI | 20.5 | 60 t/s |
| 216 | Gpt 5 6 Sol High OpenAI | 20.4 | 90 t/s |
| 217 | Kimi K3 Moonshot AI · Abierto | 19.7 | 37 t/s |
| 218 | Gpt 5 6 Sol Medium OpenAI | 19.4 | 90 t/s |
| 219 | Gpt 5 6 Sol Low OpenAI | 18.9 | 90 t/s |
| 220 | Gpt 5 5 Medium OpenAI | 18.1 | 90 t/s |
| 221 | Grok 4.3 xAI | 18.0 | 129 t/s |
| 222 | 16.7 | 80 t/s | |
| 223 | Claude Opus 4 7 Non Reasoning Anthropic | 14.8 | 50 t/s |
| 224 | DeepSeek V4 Flash DeepSeek · Abierto | 12.5 | 113 t/s |
| 225 | Kimi K2.6 Moonshot AI · Abierto | 5.3 | 40 t/s |
| 226 | Gemini 3.5 Flash-Lite Google | 5.2 | 389 t/s |
| 227 | GLM-5.2 Z.ai · Abierto | 4.4 | 164 t/s |
| 228 | Glm 5 Z.ai · Abierto | 4.4 | 60 t/s |
| 229 | Qwen3 8 2 4t A95b Alibaba · Abierto | 4.3 | 85 t/s |
| 230 | Kimi K3 Low Moonshot AI | 3.9 | 60 t/s |
| 231 | MiMo V2.5 Pro Xiaomi · Abierto | 3.3 | 65 t/s |
| 232 | Claude Opus 4.6 Anthropic | 2.4 | 50 t/s |
| 233 | Inkling Thinking Machines | 2.0 | 81 t/s |
| 234 | MiniMax M3 MiniMax | 1.4 | 93 t/s |
| 235 | Qwen3.7 Plus Alibaba | 1.1 | 54 t/s |
| 236 | 1.1 | 60 t/s | |
| 237 | GLM-5.1 Z.ai · Abierto | 0.9 | 74 t/s |
| 238 | DeepSeek V4 Pro DeepSeek · Abierto | 0.8 | 66 t/s |
| 239 | MiniMax-M2.7 MiniMax | 0.8 | 40 t/s |
| 240 | Gemini 3 5 Flash Minimal Google | 0.7 | 120 t/s |
| 241 | -0.4 | 150 t/s | |
| 242 | Claude Sonnet 5 Non Reasoning Anthropic | -0.6 | 50 t/s |
| 243 | Solar Pro4 Upstage | -0.8 | 40 t/s |
| 244 | -2.1 | 50 t/s | |
| 245 | Claude Sonnet 4.6 Anthropic | -3.5 | 50 t/s |
| 246 | Command A Plus Cohere | -4.0 | 60 t/s |
| 247 | Claude Opus 4.5 Anthropic | -4.1 | 50 t/s |
| 248 | Gemini 3 Flash Google | -4.3 | 120 t/s |
| 249 | Gpt 5 5 Non Reasoning OpenAI | -4.7 | 60 t/s |
| 250 | Glm 5 2 Non Reasoning Z.ai · Abierto | -6.6 | 40 t/s |
| 251 | -6.8 | 50 t/s | |
| 252 | Kimi K2 5 Moonshot AI | -7.3 | 60 t/s |
| 253 | Claude 4.5 Haiku Anthropic | -7.6 | 80 t/s |
| 254 | Inkling Small Thinking Machines | -8.9 | 130 t/s |
| 255 | Claude 4 Sonnet Anthropic | -9.0 | 80 t/s |
| 256 | Kimi K2 6 Non Reasoning Moonshot AI | -9.2 | 40 t/s |
| 257 | Claude 4 5 Sonnet Anthropic | -9.4 | 80 t/s |
| 258 | Claude 3 7 Sonnet Anthropic | -9.7 | 80 t/s |
| 259 | Mimo V2 5 0424 Xiaomi · Abierto | -9.8 | 60 t/s |
| 260 | GPT-5.6 Luna OpenAI | -10.3 | 175 t/s |
| 261 | Gpt 4o OpenAI | -10.5 | 90 t/s |
| 262 | Gpt 5 Mini Medium OpenAI | -11.0 | 90 t/s |
| 263 | Glm 5 Non Reasoning Z.ai · Abierto | -11.8 | 40 t/s |
| 264 | Gpt 5 2 Non Reasoning OpenAI | -12.2 | 60 t/s |
| 265 | Qwen3 5 Omni Plus Alibaba · Abierto | -13.1 | 85 t/s |
| 266 | Kimi K2 5 Non Reasoning Moonshot AI | -13.8 | 40 t/s |
| 267 | Gpt 5 4 Non Reasoning OpenAI | -15.7 | 60 t/s |
| 268 | Nemotron 3 5 Lightning NVIDIA | -17.7 | 150 t/s |
| 269 | Ling 3 0 Flash Ling | -17.9 | 60 t/s |
| 270 | Ling 3 0 Tiny Ling | -19.3 | 60 t/s |
| 271 | Gemma 4 E4b Google · Abierto | -19.7 | 120 t/s |
| 272 | Qwen3 6 27b Alibaba · Abierto | -20.0 | 85 t/s |
| 273 | Gpt 4o 2024 08 06 OpenAI | -20.9 | 90 t/s |
| 274 | Qwen3 6 35b A3b Alibaba · Abierto | -22.2 | 85 t/s |
| 275 | Qwen3 Coder 480b A35b Instruct Alibaba · Abierto | -22.2 | 85 t/s |
| 276 | Glm 5 1 Non Reasoning Z.ai · Abierto | -22.4 | 40 t/s |
| 277 | -23.2 | 60 t/s | |
| 278 | -24.9 | 60 t/s | |
| 279 | Gpt 5 Nano Medium OpenAI | -25.8 | 90 t/s |
| 280 | Kimi K2 0905 Moonshot AI | -26.6 | 60 t/s |
| 281 | Glm 4 6v Reasoning Z.ai · Abierto | -26.9 | 40 t/s |
| 282 | Kimi K2 Moonshot AI | -28.3 | 60 t/s |
| 283 | Gpt 5 Nano OpenAI | -28.7 | 90 t/s |
| 284 | Deepseek V4 Pro 0424 Non Reasoning DeepSeek · Abierto | -29.9 | 40 t/s |
| 285 | Mistral Small 4 Mistral · Abierto | -30.4 | 100 t/s |
| 286 | Llama 3 1 Instruct 8b Meta · Abierto | -30.9 | 80 t/s |
| 287 | Mistral Medium 3 Mistral · Abierto | -31.4 | 100 t/s |
| 288 | Muse Glimmer Meta | -32.8 | 80 t/s |
| 289 | -33.1 | 50 t/s | |
| 290 | Grok 3 xAI | -33.2 | 80 t/s |
| 291 | Hermes 4 Llama 3 1 405b Meta · Abierto | -33.3 | 80 t/s |
| 292 | Gpt 5 Minimal OpenAI | -33.8 | 90 t/s |
| 293 | Gpt 4 OpenAI | -33.8 | 90 t/s |
| 294 | Gpt 5 1 Non Reasoning OpenAI | -34.4 | 60 t/s |
| 295 | Hy3 Non Reasoning Tencent | -34.9 | 40 t/s |
| 296 | Ernie 4 5 300b A47b Baidu | -35.1 | 60 t/s |
| 297 | Nova Premier Amazon | -35.4 | 60 t/s |
| 298 | Mistral Medium 3.5 Mistral | -36.8 | 60 t/s |
| 299 | -37.5 | 60 t/s | |
| 300 | Mimo V2 5 Pro Non Reasoning Xiaomi · Abierto | -37.8 | 40 t/s |
| 301 | Qwen3 5 397b A17b Non Reasoning Alibaba · Abierto | -37.9 | 55 t/s |
| 302 | Gpt 4 1 OpenAI | -39.6 | 90 t/s |
| 303 | Mistral Large 3 Mistral · Abierto | -39.6 | 100 t/s |
| 304 | -40.3 | 40 t/s | |
| 305 | Deepseek V3 0324 DeepSeek · Abierto | -40.7 | 60 t/s |
| 306 | Deepseek V3 DeepSeek · Abierto | -40.7 | 60 t/s |
| 307 | Llama 3 1 Nemotron Instruct 70b Meta · Abierto | -40.8 | 80 t/s |
| 308 | Gemma 4 E4b Non Reasoning Google · Abierto | -41.0 | 80 t/s |
| 309 | -41.5 | 150 t/s | |
| 310 | -41.5 | 100 t/s | |
| 311 | Llama 4 Maverick Meta · Abierto | -41.8 | 80 t/s |
| 312 | Nova Lite Amazon | -42.2 | 60 t/s |
| 313 | -42.6 | 50 t/s | |
| 314 | -42.6 | 120 t/s | |
| 315 | Qwen3 Max Preview Alibaba · Abierto | -42.8 | 85 t/s |
| 316 | Llama 3 1 Instruct 70b Meta · Abierto | -43.1 | 80 t/s |
| 317 | Qwen3 Max Alibaba · Abierto | -43.5 | 85 t/s |
| 318 | Deepseek V4 Flash Non Reasoning DeepSeek · Abierto | -44.0 | 40 t/s |
| 319 | Qwen3 235b A22b Instruct 2507 Alibaba · Abierto | -44.0 | 85 t/s |
| 320 | -44.6 | 50 t/s | |
| 321 | -45.4 | 60 t/s | |
| 322 | Llama Nemotron Super 49b V1 5 Meta · Abierto | -46.2 | 80 t/s |
| 323 | Mistral Medium 3 1 Mistral · Abierto | -46.3 | 100 t/s |
| 324 | Glm 4 5v Reasoning Z.ai · Abierto | -46.3 | 40 t/s |
| 325 | Deepseek R1 Distill Llama 70b DeepSeek · Abierto | -46.4 | 40 t/s |
| 326 | Devstral 2 Mistral | -46.7 | 100 t/s |
| 327 | DeepSeek V3.2 DeepSeek · Abierto | -46.9 | 60 t/s |
| 328 | Glm 4 7 Non Reasoning Z.ai · Abierto | -47.3 | 40 t/s |
| 329 | Llama Nemotron Super 49b V1 5 Reasoning Meta · Abierto | -47.6 | 50 t/s |
| 330 | Qwen3 5 27b Non Reasoning Alibaba · Abierto | -47.7 | 55 t/s |
| 331 | Nova Pro Amazon | -47.7 | 60 t/s |
| 332 | Gemma 4 31B Google · Abierto | -47.9 | 35 t/s |
| 333 | Deepseek V3 2 0925 DeepSeek · Abierto | -48.2 | 60 t/s |
| 334 | Command A Cohere | -48.3 | 60 t/s |
| 335 | Mistral Small 4 Non Reasoning Mistral · Abierto | -48.5 | 60 t/s |
| 336 | Nova Micro Amazon | -48.5 | 60 t/s |
| 337 | North Mini Code North | -48.6 | 60 t/s |
| 338 | Hermes 4 Llama 3 1 70b Meta · Abierto | -48.6 | 80 t/s |
| 339 | Claude 3 Haiku Anthropic | -48.6 | 80 t/s |
| 340 | Qwen3 14b Instruct Reasoning Alibaba · Abierto | -48.9 | 55 t/s |
| 341 | GPT-oss-120B OpenAI · Abierto | -49.2 | 240 t/s |
| 342 | Hermes 4 Llama 3 1 70b Reasoning Meta · Abierto | -49.4 | 50 t/s |
| 343 | Nova 2.0 Pro Preview Amazon · preview | -49.5 | 40 t/s |
| 344 | Qwen3 32b Instruct Reasoning Alibaba · Abierto | -50.4 | 55 t/s |
| 345 | Ling 2 6 1t Ling | -50.6 | 60 t/s |
| 346 | Mistral Small 3 2 Mistral · Abierto | -50.6 | 100 t/s |
| 347 | Mistral Small 3 1 Mistral · Abierto | -50.7 | 100 t/s |
| 348 | Qwen3 Coder 30b A3b Instruct Alibaba · Abierto | -50.7 | 85 t/s |
| 349 | Gemma 4 26b A4b Google · Abierto | -50.8 | 120 t/s |
| 350 | Qwen3 Next 80b A3b Reasoning Alibaba · Abierto | -51.4 | 55 t/s |
| 351 | Gemma 4 31b Non Reasoning Google · Abierto | -51.7 | 80 t/s |
| 352 | Qwen3 Vl 8b Instruct Alibaba · Abierto | -51.9 | 85 t/s |
| 353 | Llama 4 Scout Meta · Abierto | -52.1 | 80 t/s |
| 354 | Qwen3 235b A22b Instruct Alibaba · Abierto | -52.4 | 85 t/s |
| 355 | Qwen3 Vl 235b A22b Instruct Alibaba · Abierto | -52.7 | 85 t/s |
| 356 | Gemma 4 12b Non Reasoning Google · Abierto | -52.8 | 80 t/s |
| 357 | Qwen3 6 27b Non Reasoning Alibaba · Abierto | -53.2 | 55 t/s |
| 358 | Llama 3 3 Instruct 70b Meta · Abierto | -54.2 | 80 t/s |
| 359 | Llama 3 Instruct 70b Meta · Abierto | -54.3 | 80 t/s |
| 360 | Qwen3 5 122b A10b Non Reasoning Alibaba · Abierto | -55.0 | 55 t/s |
| 361 | Phi 4 Microsoft · Abierto | -55.7 | 60 t/s |
| 362 | -57.2 | 150 t/s | |
| 363 | Devstral Small 2 Mistral | -57.7 | 100 t/s |
| 364 | Qwen3 Next 80b A3b Instruct Alibaba · Abierto | -59.5 | 85 t/s |
| 365 | Qwen3 6 35b A3b Non Reasoning Alibaba · Abierto | -60.0 | 55 t/s |
| 366 | Granite 4 0 H Small IBM · Abierto | -61.0 | 60 t/s |
| 367 | Phi 4 Mini Microsoft · Abierto | -61.1 | 60 t/s |
| 368 | Apertus 70b Instruct Swiss AI | -61.5 | 60 t/s |
| 369 | Gemma 4 26b A4b Non Reasoning Google · Abierto | -62.3 | 80 t/s |
| 370 | Qwen3 Coder Next Alibaba · Abierto | -62.4 | 85 t/s |
| 371 | Llama 3 2 Instruct 11b Vision Meta · Abierto | -62.6 | 80 t/s |
| 372 | Qwen3 5 35b A3b Non Reasoning Alibaba · Abierto | -63.0 | 55 t/s |
| 373 | Qwen3 Vl 32b Instruct Alibaba · Abierto | -63.4 | 85 t/s |
| 374 | Qwen3 Vl 30b A3b Instruct Alibaba · Abierto | -63.4 | 85 t/s |
| 375 | Ministral 3 3b Mistral | -64.0 | 100 t/s |
| 376 | Gpt 5 Nano Minimal OpenAI | -64.1 | 90 t/s |
| 377 | Mistral 7b Instruct Mistral · Abierto | -64.9 | 100 t/s |
| 378 | Qwen3 30b A3b Instruct Alibaba · Abierto | -66.0 | 85 t/s |
| 379 | Ling 2 6 Flash Ling | -66.1 | 60 t/s |
| 380 | Qwen3 30b A3b 2507 Alibaba · Abierto | -66.1 | 85 t/s |
| 381 | Ministral 3 14b Mistral | -66.4 | 100 t/s |
| 382 | Qwen3 14b Instruct Alibaba · Abierto | -66.7 | 85 t/s |
| 383 | Gemma 3 27b Google · Abierto | -67.2 | 120 t/s |
| 384 | Glm 4 7 Flash Non Reasoning Z.ai · Abierto | -68.9 | 40 t/s |
| 385 | Ministral 3 8b Mistral | -68.9 | 100 t/s |
| 386 | Gemma 3 1b Google · Abierto | -75.4 | 120 t/s |
| 387 | Gemma 3 4b Google · Abierto | -82.9 | 120 t/s |
| 388 | Tiny Aya Global Cohere | -84.3 | 60 t/s |
Podium Score es una mezcla ponderada de cinco clasificaciones. Consulta la Metodología para más detalles.