💰 Value Rankings
Intelligence per dollar: blended API pricing and cost-per-task, inverted so cheaper is better.
🏆 Overall673💻 Coding392📐 Math32🧠 Reasoning391🤖 Agentic31📚 Knowledge388👁️ Multimodal370🛠️ Coding Agents40🔬 Scientific Reasoning45🌍 Multilingual12✍️ Writing53💬 Chat & Assistants53🔭 Research48🔧 Tool Calling31📜 Long Context55⚡ Speed378💰 Value377🔓 Open Weights153
377 models
| # | Model | Podium Score | Speed |
|---|---|---|---|
| 01 | MiMo V2.5 Pro Xiaomi · Open | 100.0 | 65 t/s |
| 02 | DeepSeek V4 Flash DeepSeek · Open | 100.0 | 113 t/s |
| 03 | Hunyuan Hy3 Tencent | 99.9 | 69 t/s |
| 4 | GPT-5.6 Luna OpenAI | 99.8 | 175 t/s |
| 5 | DeepSeek V4 Pro DeepSeek · Open | 99.8 | 66 t/s |
| 6 | Inkling Small Thinking Machines | 99.6 | 130 t/s |
| 7 | Gemini 3.5 Flash-Lite Google | 99.4 | 389 t/s |
| 8 | MiniMax M3 MiniMax | 99.0 | 93 t/s |
| 9 | Kimi K2.7 Code Moonshot AI · Open | 98.3 | 39 t/s |
| 10 | Qwen3.7 Plus Alibaba | 98.2 | 54 t/s |
| 11 | Qwen3.5 122B-A10B Alibaba · Open | 97.9 | 126 t/s |
| 12 | Muse Spark 1.1 Meta | 97.6 | 208 t/s |
| 13 | Grok 4.3 xAI | 97.6 | 129 t/s |
| 14 | Gemini 3.1 Pro Google | 97.2 | 136 t/s |
| 15 | Grok 4.5 xAI | 97.0 | 59 t/s |
| 16 | Qwen3.6 Plus Alibaba | 97.0 | 55 t/s |
| 17 | Qwen3.5 397B-A17B Alibaba · Open | 97.0 | 71 t/s |
| 18 | Nemotron 3 Ultra 550B NVIDIA · Open | 96.8 | 142 t/s |
| 19 | GPT-5.6 Terra OpenAI | 95.6 | 127 t/s |
| 20 | Gemini 3.6 Flash Google | 95.1 | 233 t/s |
| 21 | GLM-5.2 Z.ai · Open | 95.0 | 164 t/s |
| 22 | Gemini 3.5 Flash Google | 93.9 | 267 t/s |
| 23 | Kimi K3 Moonshot AI · Open | 92.4 | 37 t/s |
| 24 | Qwen3.7 Max Alibaba | 90.4 | 203 t/s |
| 25 | GPT-5.6 Sol OpenAI | 89.0 | 72 t/s |
| 26 | Claude Sonnet 5 Anthropic | 84.5 | 96 t/s |
| 27 | Nex N2 Pro Nex AGI | 81.9 | 138 t/s |
| 28 | Claude Opus 5 Anthropic | 78.8 | 60 t/s |
| 29 | Claude Fable 5 Anthropic | 71.4 | 71 t/s |
| 30 | Inkling Thinking Machines | 70.1 | 81 t/s |
| 31 | GLM-5.1 Z.ai · Open | 67.0 | 74 t/s |
| 32 | Claude Sonnet 4.6 (max) Anthropic | 67.0 | 50 t/s |
| 33 | MiniMax-M2.7 MiniMax | 64.0 | 40 t/s |
| 34 | Gemini 3 7 Flash Google | 60.0 | 120 t/s |
| 35 | Gemini 3 7 Flash Medium Google | 60.0 | 120 t/s |
| 36 | Gemini 3 7 Flash Low Google | 60.0 | 120 t/s |
| 37 | Gpt 5 4 Mini OpenAI | 60.0 | 90 t/s |
| 38 | Deepseek V4 Flash 0420 DeepSeek · Open | 60.0 | 60 t/s |
| 39 | Gpt 5 4 Nano OpenAI | 60.0 | 90 t/s |
| 40 | Deepseek V4 Flash 0420 High DeepSeek · Open | 60.0 | 60 t/s |
| 41 | Gemini 3 Flash Reasoning Google | 60.0 | 80 t/s |
| 42 | Ling 3 0 Flash Ling | 60.0 | 60 t/s |
| 43 | Gemini 3 5 Flash Minimal Google | 60.0 | 120 t/s |
| 44 | Gemini 3 Pro Low Google | 60.0 | 120 t/s |
| 45 | Minimax M2 1 MiniMax · Open | 60.0 | 60 t/s |
| 46 | Mimo V2 Flash Reasoning Xiaomi · Open | 60.0 | 40 t/s |
| 47 | Gpt 5 Mini Medium OpenAI | 60.0 | 90 t/s |
| 48 | Gpt 5 1 Codex Mini OpenAI | 60.0 | 90 t/s |
| 49 | Step 3 7 Flash StepFun | 60.0 | 60 t/s |
| 50 | Gpt 5 4 Nano Medium OpenAI | 60.0 | 90 t/s |
| 51 | Gpt 5 4 Mini Medium OpenAI | 60.0 | 90 t/s |
| 52 | Deepseek V4 Flash Non Reasoning DeepSeek · Open | 60.0 | 40 t/s |
| 53 | Minimax M2 MiniMax · Open | 60.0 | 60 t/s |
| 54 | O4 Mini OpenAI | 60.0 | 60 t/s |
| 55 | Step 3 5 Flash 0202 StepFun | 60.0 | 60 t/s |
| 56 | Gpt 5 Mini OpenAI | 60.0 | 90 t/s |
| 57 | 60.0 | 50 t/s | |
| 58 | Gemini 2 5 Pro 05 06 Google | 60.0 | 120 t/s |
| 59 | Nova 2 0 Lite Reasoning Amazon | 60.0 | 40 t/s |
| 60 | 60.0 | 80 t/s | |
| 61 | North Mini Code North | 60.0 | 60 t/s |
| 62 | Gpt 5 Nano OpenAI | 60.0 | 90 t/s |
| 63 | Gpt 5 Nano Medium OpenAI | 60.0 | 90 t/s |
| 64 | 60.0 | 40 t/s | |
| 65 | O3 Mini OpenAI | 60.0 | 60 t/s |
| 66 | Qwen3 5 Omni Flash Alibaba · Open | 60.0 | 85 t/s |
| 67 | 60.0 | 40 t/s | |
| 68 | Minimax M1 80k MiniMax · Open | 60.0 | 60 t/s |
| 69 | 60.0 | 60 t/s | |
| 70 | Gpt 5 Minimal OpenAI | 60.0 | 90 t/s |
| 71 | 60.0 | 60 t/s | |
| 72 | O3 Mini High OpenAI | 60.0 | 60 t/s |
| 73 | Glm 4 7 Flash Non Reasoning Z.ai · Open | 60.0 | 40 t/s |
| 74 | 60.0 | 80 t/s | |
| 75 | 60.0 | 150 t/s | |
| 76 | Gpt 4 1 Mini OpenAI | 60.0 | 90 t/s |
| 77 | 60.0 | 100 t/s | |
| 78 | Gpt 5 Mini Minimal OpenAI | 60.0 | 90 t/s |
| 79 | Ling 2 6 Flash Ling | 60.0 | 60 t/s |
| 80 | 60.0 | 120 t/s | |
| 81 | Nova 2 0 Lite Amazon | 60.0 | 60 t/s |
| 82 | 60.0 | 80 t/s | |
| 83 | Ministral 3 14b Mistral | 60.0 | 100 t/s |
| 84 | 60.0 | 120 t/s | |
| 85 | Qwen3 Vl 8b Reasoning Alibaba · Open | 60.0 | 55 t/s |
| 86 | Qwen3 14b Instruct Reasoning Alibaba · Open | 60.0 | 55 t/s |
| 87 | Gpt 4 1 Nano OpenAI | 60.0 | 90 t/s |
| 88 | Ministral 3 8b Mistral | 60.0 | 100 t/s |
| 89 | 60.0 | 100 t/s | |
| 90 | 60.0 | 100 t/s | |
| 91 | Qwen3 8b Instruct Reasoning Alibaba · Open | 60.0 | 55 t/s |
| 92 | Qwen3 Vl 8b Instruct Alibaba · Open | 60.0 | 85 t/s |
| 93 | Gpt 5 Nano Minimal OpenAI | 60.0 | 90 t/s |
| 94 | Llama 3 1 Instruct 8b Meta · Open | 60.0 | 80 t/s |
| 95 | 60.0 | 150 t/s | |
| 96 | 60.0 | 150 t/s | |
| 97 | Ministral 3 3b Mistral | 60.0 | 100 t/s |
| 98 | Qwen3 14b Instruct Alibaba · Open | 60.0 | 85 t/s |
| 99 | Gemini 2 5 Flash Lite Google | 60.0 | 120 t/s |
| 100 | Gpt 4o Mini OpenAI | 60.0 | 90 t/s |
| 101 | Nova Lite Amazon | 60.0 | 60 t/s |
| 102 | Reka Flash Reka | 60.0 | 60 t/s |
| 103 | Solar Mini Upstage | 60.0 | 60 t/s |
| 104 | Phi 4 Mini Microsoft · Open | 60.0 | 60 t/s |
| 105 | Qwen3 8b Instruct Alibaba · Open | 60.0 | 85 t/s |
| 106 | 60.0 | 150 t/s | |
| 107 | Llama 2 Chat 7b Meta · Open | 60.0 | 80 t/s |
| 108 | Reka Flash 3 Reka | 60.0 | 60 t/s |
| 109 | Olmo 3 7b Instruct AI2 · Open | 60.0 | 60 t/s |
| 110 | Jamba 1 5 Mini AI21 Labs | 60.0 | 60 t/s |
| 111 | Mistral 7b Instruct Mistral · Open | 60.0 | 100 t/s |
| 112 | Granite 3 3 8b Instruct IBM · Open | 60.0 | 60 t/s |
| 113 | Apertus 8b Instruct Swiss AI | 60.0 | 60 t/s |
| 114 | Gemma 3 1b Google · Open | 60.0 | 120 t/s |
| 115 | Llama 3 Instruct 8b Meta · Open | 60.0 | 80 t/s |
| 116 | Async Flash V1 5 Async | 60.0 | 60 t/s |
| 117 | Gemini Google | 60.0 | 120 t/s |
| 118 | 60.0 | 120 t/s | |
| 119 | Gemini_nano Banana Pro Google | 60.0 | 120 t/s |
| 120 | Minimax H3 MiniMax · Open | 60.0 | 60 t/s |
| 121 | Mistral Medium 3.5 Mistral | 51.0 | 60 t/s |
| 122 | Claude 4.5 Haiku Anthropic | 48.0 | 80 t/s |
| 123 | Nova 2.0 Pro Preview Amazon · preview | 46.0 | 40 t/s |
| 124 | Claude Mythos 5 Anthropic · preview | 45.0 | 50 t/s |
| 125 | GPT-5.5 Pro OpenAI | 45.0 | 60 t/s |
| 126 | GPT-oss-120B OpenAI · Open | 43.0 | 240 t/s |
| 127 | Claude Opus 5 Xhigh Anthropic | 40.0 | 80 t/s |
| 128 | Claude Opus 5 High Anthropic | 40.0 | 80 t/s |
| 129 | Grok 4 xAI | 40.0 | 80 t/s |
| 130 | Grok 4 6 xAI | 40.0 | 80 t/s |
| 131 | Claude Opus 5 Medium Anthropic | 40.0 | 80 t/s |
| 132 | Qwen3 8 2 4t A95b Alibaba · Open | 40.0 | 85 t/s |
| 133 | Gpt 5 6 Sol High OpenAI | 40.0 | 90 t/s |
| 134 | Muse Spark 1 2 Meta | 40.0 | 80 t/s |
| 135 | Gpt 5 6 Sol Medium OpenAI | 40.0 | 90 t/s |
| 136 | Glm 5 Z.ai · Open | 40.0 | 60 t/s |
| 137 | Claude Opus 5 Low Anthropic | 40.0 | 80 t/s |
| 138 | Gpt 5 5 Medium OpenAI | 40.0 | 90 t/s |
| 139 | Gpt 5 6 Sol Low OpenAI | 40.0 | 90 t/s |
| 140 | Gpt 5 6 Terra High OpenAI | 40.0 | 90 t/s |
| 141 | Kimi K3 Low Moonshot AI | 40.0 | 60 t/s |
| 142 | Claude Sonnet 4 6 Adaptive Anthropic | 40.0 | 80 t/s |
| 143 | Gpt 5 6 Luna High OpenAI | 40.0 | 90 t/s |
| 144 | Gpt 5 6 Terra Medium OpenAI | 40.0 | 90 t/s |
| 145 | Deepseek V4 Pro 0424 DeepSeek · Open | 40.0 | 60 t/s |
| 146 | Claude Opus 4 6 Adaptive Anthropic | 40.0 | 80 t/s |
| 147 | Gpt 5 5 Low OpenAI | 40.0 | 90 t/s |
| 148 | Claude Opus 4 7 Non Reasoning Anthropic | 40.0 | 50 t/s |
| 149 | Deepseek V4 Pro 0424 High DeepSeek · Open | 40.0 | 60 t/s |
| 150 | Claude Sonnet 5 Non Reasoning Anthropic | 40.0 | 50 t/s |
| 151 | Claude Opus 4 5 Thinking Anthropic | 40.0 | 50 t/s |
| 152 | 40.0 | 60 t/s | |
| 153 | Solar Pro4 Upstage | 40.0 | 40 t/s |
| 154 | Gpt 5 6 Terra Low OpenAI | 40.0 | 90 t/s |
| 155 | Gpt 5 2 Codex OpenAI | 40.0 | 90 t/s |
| 156 | Qwen3 6 Max Alibaba · Open | 40.0 | 85 t/s |
| 157 | 40.0 | 80 t/s | |
| 158 | Gpt 5 4 Low OpenAI | 40.0 | 90 t/s |
| 159 | Agnes 2 5 Pro Alpha Agnes | 40.0 | 60 t/s |
| 160 | Gpt 5 2 Medium OpenAI | 40.0 | 90 t/s |
| 161 | Gpt 5 6 Luna Medium OpenAI | 40.0 | 90 t/s |
| 162 | 40.0 | 150 t/s | |
| 163 | Deepseek V3 2 Reasoning DeepSeek · Open | 40.0 | 40 t/s |
| 164 | Grok 4 20 xAI | 40.0 | 80 t/s |
| 165 | Mimo V2 5 0424 Xiaomi · Open | 40.0 | 60 t/s |
| 166 | Qwen3 6 27b Alibaba · Open | 40.0 | 85 t/s |
| 167 | Claude 4 5 Sonnet Thinking Anthropic | 40.0 | 50 t/s |
| 168 | 40.0 | 80 t/s | |
| 169 | Gpt 5 Codex OpenAI | 40.0 | 90 t/s |
| 170 | 40.0 | 80 t/s | |
| 171 | Glm 5 1 Non Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 172 | Grok 4 3 Low xAI | 40.0 | 80 t/s |
| 173 | Kimi K2 5 Moonshot AI | 40.0 | 60 t/s |
| 174 | Gpt 5 5 Non Reasoning OpenAI | 40.0 | 60 t/s |
| 175 | Gpt 5 1 Codex OpenAI | 40.0 | 90 t/s |
| 176 | Kimi K2 6 Non Reasoning Moonshot AI | 40.0 | 40 t/s |
| 177 | 40.0 | 50 t/s | |
| 178 | Muse Glimmer Meta | 40.0 | 80 t/s |
| 179 | Glm 5 2 Non Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 180 | 40.0 | 60 t/s | |
| 181 | Gpt 5 Medium OpenAI | 40.0 | 90 t/s |
| 182 | Claude 4 1 Opus Thinking Anthropic | 40.0 | 50 t/s |
| 183 | Hy3 Preview Tencent | 40.0 | 60 t/s |
| 184 | Gpt 5 5 Instant 05 26 OpenAI | 40.0 | 90 t/s |
| 185 | Gpt 5 6 Luna Low OpenAI | 40.0 | 90 t/s |
| 186 | Kimi K2 Thinking Moonshot AI | 40.0 | 40 t/s |
| 187 | O3 Pro OpenAI | 40.0 | 60 t/s |
| 188 | Glm 5 Non Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 189 | Claude 4 5 Haiku Reasoning Anthropic | 40.0 | 50 t/s |
| 190 | Qwen3 5 397b A17b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 191 | Qwen3 6 35b A3b Alibaba · Open | 40.0 | 85 t/s |
| 192 | 40.0 | 40 t/s | |
| 193 | 40.0 | 150 t/s | |
| 194 | Deepseek V4 Pro 0424 Non Reasoning DeepSeek · Open | 40.0 | 40 t/s |
| 195 | Gpt 5 Low OpenAI | 40.0 | 90 t/s |
| 196 | Claude 4 Opus Thinking Anthropic | 40.0 | 50 t/s |
| 197 | Qwen3 5 Omni Plus Alibaba · Open | 40.0 | 85 t/s |
| 198 | Qwen3 6 27b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 199 | Ring 2 6 1t WeRide | 40.0 | 60 t/s |
| 200 | Deepseek V3 1 Terminus Reasoning DeepSeek · Open | 40.0 | 40 t/s |
| 201 | Kimi K2 5 Non Reasoning Moonshot AI | 40.0 | 40 t/s |
| 202 | Qwen3 5 27b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 203 | Claude 4 5 Sonnet Anthropic | 40.0 | 80 t/s |
| 204 | Claude 4 Sonnet Thinking Anthropic | 40.0 | 50 t/s |
| 205 | Glm 4 6 Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 206 | Gpt 5 5 Instant 06 26 OpenAI | 40.0 | 90 t/s |
| 207 | Claude 4 1 Opus Anthropic | 40.0 | 80 t/s |
| 208 | Mimo V2 5 Pro Non Reasoning Xiaomi · Open | 40.0 | 40 t/s |
| 209 | Gpt 5 4 Non Reasoning OpenAI | 40.0 | 60 t/s |
| 210 | Qwen3 5 122b A10b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 211 | 40.0 | 50 t/s | |
| 212 | Glm 4 7 Non Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 213 | 40.0 | 60 t/s | |
| 214 | Hy3 Non Reasoning Tencent | 40.0 | 40 t/s |
| 215 | Ling 2 6 1t Ling | 40.0 | 60 t/s |
| 216 | Gpt 5 2 Non Reasoning OpenAI | 40.0 | 60 t/s |
| 217 | Gemma 4 26b A4b Google · Open | 40.0 | 120 t/s |
| 218 | Claude 4 Opus Anthropic | 40.0 | 80 t/s |
| 219 | Claude 4 Sonnet Anthropic | 40.0 | 80 t/s |
| 220 | Deepseek V3 2 Reasoning 0925 DeepSeek · Open | 40.0 | 40 t/s |
| 221 | Qwen3 Max Thinking Preview Alibaba · Open | 40.0 | 55 t/s |
| 222 | 40.0 | 50 t/s | |
| 223 | Qwen3 6 35b A3b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 224 | Ling 3 0 Tiny Ling | 40.0 | 60 t/s |
| 225 | Qwen3 5 35b A3b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 226 | Kimi K2 0905 Moonshot AI | 40.0 | 60 t/s |
| 227 | Claude 3 7 Sonnet Anthropic | 40.0 | 80 t/s |
| 228 | O1 OpenAI | 40.0 | 60 t/s |
| 229 | Nemotron 3 5 Lightning NVIDIA | 40.0 | 150 t/s |
| 230 | 40.0 | 50 t/s | |
| 231 | Command A Plus Cohere | 40.0 | 60 t/s |
| 232 | Gemma 4 31b Non Reasoning Google · Open | 40.0 | 80 t/s |
| 233 | Gemma 4 12b Google · Open | 40.0 | 120 t/s |
| 234 | 40.0 | 50 t/s | |
| 235 | Mercury 2 Inception | 40.0 | 60 t/s |
| 236 | Qwen3 5 9b Alibaba · Open | 40.0 | 85 t/s |
| 237 | Deepseek V3 2 0925 DeepSeek · Open | 40.0 | 60 t/s |
| 238 | Apriel V1 5 15b Thinker ServiceNow | 40.0 | 60 t/s |
| 239 | 40.0 | 40 t/s | |
| 240 | Qwen3 Coder Next Alibaba · Open | 40.0 | 85 t/s |
| 241 | Deepseek V3 1 Reasoning DeepSeek · Open | 40.0 | 40 t/s |
| 242 | Qwen3 Vl 235b A22b Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 243 | Apriel V1 6 15b Thinker ServiceNow | 40.0 | 60 t/s |
| 244 | Gpt 5 1 Non Reasoning OpenAI | 40.0 | 60 t/s |
| 245 | Qwen3 5 9b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 246 | Gemma 4 26b A4b Non Reasoning Google · Open | 40.0 | 80 t/s |
| 247 | Qwen3 5 4b Alibaba · Open | 40.0 | 85 t/s |
| 248 | Qwen3 235b A22b Instruct 2507 Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 249 | 40.0 | 40 t/s | |
| 250 | Kimi K2 Moonshot AI | 40.0 | 60 t/s |
| 251 | Mistral Small 4 Mistral · Open | 40.0 | 100 t/s |
| 252 | Gpt 4 1 OpenAI | 40.0 | 90 t/s |
| 253 | Qwen3 Max Preview Alibaba · Open | 40.0 | 85 t/s |
| 254 | Devstral 2 Mistral | 40.0 | 100 t/s |
| 255 | O1 Pro OpenAI | 40.0 | 60 t/s |
| 256 | Deepseek R1 0120 DeepSeek · Open | 40.0 | 40 t/s |
| 257 | Grok 3 xAI | 40.0 | 80 t/s |
| 258 | Seed Oss 36b Instruct ByteDance | 40.0 | 60 t/s |
| 259 | Qwen3 235b A22b Instruct 2507 Alibaba · Open | 40.0 | 85 t/s |
| 260 | Trinity Large Thinking Trinity | 40.0 | 40 t/s |
| 261 | Qwen3 Coder 480b A35b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 262 | Qwen3 Vl 32b Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 263 | Devstral Small 2 Mistral | 40.0 | 100 t/s |
| 264 | O1 Preview OpenAI | 40.0 | 60 t/s |
| 265 | Glm 4 6v Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 266 | Qwen3 Next 80b A3b Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 267 | 40.0 | 40 t/s | |
| 268 | Grok 4 Fast xAI | 40.0 | 80 t/s |
| 269 | Qwen3 5 4b Non Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 270 | Mistral Large 3 Mistral · Open | 40.0 | 100 t/s |
| 271 | Deepseek V3 0324 DeepSeek · Open | 40.0 | 60 t/s |
| 272 | Gpt Oss 120b Low OpenAI | 40.0 | 90 t/s |
| 273 | Mistral Small 3 1 Mistral · Open | 40.0 | 100 t/s |
| 274 | Mistral Medium 3 1 Mistral · Open | 40.0 | 100 t/s |
| 275 | Qwen3 30b A3b 2507 Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 276 | Llama 4 Maverick Meta · Open | 40.0 | 80 t/s |
| 277 | Gpt Oss 20b Low OpenAI | 40.0 | 90 t/s |
| 278 | Qwen3 Vl 235b A22b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 279 | Deepseek V3 DeepSeek · Open | 40.0 | 60 t/s |
| 280 | Qwen3 Next 80b A3b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 281 | Qwen3 Coder 30b A3b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 282 | Qwen3 235b A22b Instruct Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 283 | Qwen3 Vl 30b A3b Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 284 | Qwq 32b Alibaba | 40.0 | 85 t/s |
| 285 | Gemma 4 12b Non Reasoning Google · Open | 40.0 | 80 t/s |
| 286 | Nova Premier Amazon | 40.0 | 60 t/s |
| 287 | Mistral Medium 3 Mistral · Open | 40.0 | 100 t/s |
| 288 | Celeris 1 Celeris | 40.0 | 60 t/s |
| 289 | Llama Nemotron Super 49b V1 5 Reasoning Meta · Open | 40.0 | 50 t/s |
| 290 | Mistral Small 4 Non Reasoning Mistral · Open | 40.0 | 60 t/s |
| 291 | Gemma 4 E4b Google · Open | 40.0 | 120 t/s |
| 292 | Claude 3 Opus Anthropic | 40.0 | 80 t/s |
| 293 | Qwen3 32b Instruct Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 294 | Gpt 4o OpenAI | 40.0 | 90 t/s |
| 295 | Qwen3 Vl 32b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 296 | Qwen3 235b A22b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 297 | Mistral Small 3 2 Mistral · Open | 40.0 | 100 t/s |
| 298 | Nova 2 0 Omni Amazon | 40.0 | 60 t/s |
| 299 | Llama 4 Scout Meta · Open | 40.0 | 80 t/s |
| 300 | Hermes 4 Llama 3 1 70b Reasoning Meta · Open | 40.0 | 50 t/s |
| 301 | Qwen3 Vl 30b A3b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 302 | Claude 35 Sonnet Anthropic | 40.0 | 80 t/s |
| 303 | Deepseek R1 Distill Llama 70b DeepSeek · Open | 40.0 | 40 t/s |
| 304 | Qwen3 Omni 30b A3b Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 305 | Gpt 4o 2024 08 06 OpenAI | 40.0 | 90 t/s |
| 306 | Llama 3 3 Instruct 70b Meta · Open | 40.0 | 80 t/s |
| 307 | Qwen3 30b A3b Instruct Reasoning Alibaba · Open | 40.0 | 55 t/s |
| 308 | Glm 4 5v Reasoning Z.ai · Open | 40.0 | 40 t/s |
| 309 | Ernie 4 5 300b A47b Baidu | 40.0 | 60 t/s |
| 310 | Llama 3 1 Nemotron Ultra 253b V1 Reasoning Meta · Open | 40.0 | 50 t/s |
| 311 | Qwen3 30b A3b 2507 Alibaba · Open | 40.0 | 85 t/s |
| 312 | Hermes 4 Llama 3 1 405b Reasoning Meta · Open | 40.0 | 50 t/s |
| 313 | Gemma 4 E4b Non Reasoning Google · Open | 40.0 | 80 t/s |
| 314 | Hermes 4 Llama 3 1 405b Meta · Open | 40.0 | 80 t/s |
| 315 | Llama Nemotron Super 49b V1 5 Meta · Open | 40.0 | 80 t/s |
| 316 | Qwen3 32b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 317 | Gpt 4o 2024 05 13 OpenAI | 40.0 | 90 t/s |
| 318 | Claude 35 Sonnet June 24 Anthropic | 40.0 | 80 t/s |
| 319 | Gpt 4 Turbo OpenAI | 40.0 | 90 t/s |
| 320 | Command A Cohere | 40.0 | 60 t/s |
| 321 | Nova Pro Amazon | 40.0 | 60 t/s |
| 322 | Gemma 3 27b Google · Open | 40.0 | 120 t/s |
| 323 | Llama 3 1 Nemotron Instruct 70b Meta · Open | 40.0 | 80 t/s |
| 324 | Mistral Large 2407 Mistral · Open | 40.0 | 100 t/s |
| 325 | Gpt 4 OpenAI | 40.0 | 90 t/s |
| 326 | Hermes 4 Llama 3 1 70b Meta · Open | 40.0 | 80 t/s |
| 327 | Mistral Small 3 Mistral · Open | 40.0 | 100 t/s |
| 328 | Qwen3 30b A3b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 329 | Llama 3 1 Instruct 70b Meta · Open | 40.0 | 80 t/s |
| 330 | Qwen Turbo Alibaba · Open | 40.0 | 85 t/s |
| 331 | Gemma 3 12b Google · Open | 40.0 | 120 t/s |
| 332 | Granite 4 0 H Small IBM · Open | 40.0 | 60 t/s |
| 333 | Hermes 3 Llama 3 1 70b Meta · Open | 40.0 | 80 t/s |
| 334 | Jamba 1 5 Large AI21 Labs | 40.0 | 60 t/s |
| 335 | Qwen3 Omni 30b A3b Instruct Alibaba · Open | 40.0 | 85 t/s |
| 336 | Phi 4 Microsoft · Open | 40.0 | 60 t/s |
| 337 | Nova Micro Amazon | 40.0 | 60 t/s |
| 338 | Mistral Small Mistral · Open | 40.0 | 100 t/s |
| 339 | Mistral Large Mistral · Open | 40.0 | 100 t/s |
| 340 | Claude 3 Haiku Anthropic | 40.0 | 80 t/s |
| 341 | Gpt 35 Turbo OpenAI | 40.0 | 90 t/s |
| 342 | Mistral Medium Mistral · Open | 40.0 | 100 t/s |
| 343 | Mistral Small 2402 Mistral · Open | 40.0 | 100 t/s |
| 344 | Llama 3 Instruct 70b Meta · Open | 40.0 | 80 t/s |
| 345 | Llama 3 2 Instruct 11b Vision Meta · Open | 40.0 | 80 t/s |
| 346 | Command R Plus 04 2024 Cohere | 40.0 | 60 t/s |
| 347 | Apertus 70b Instruct Swiss AI | 40.0 | 60 t/s |
| 348 | Mixtral 8x7b Instruct Mistral | 40.0 | 100 t/s |
| 349 | Command R 03 2024 Cohere | 40.0 | 60 t/s |
| 350 | Gemma 3 4b Google · Open | 40.0 | 120 t/s |
| 351 | Gemma 3n E2b Google · Open | 40.0 | 120 t/s |
| 352 | Gemma 3n E4b Google · Open | 40.0 | 120 t/s |
| 353 | Tiny Aya Global Cohere | 40.0 | 60 t/s |
| 354 | Amazon Nova Amazon | 40.0 | 60 t/s |
| 355 | Async Pro V1 0 Async | 40.0 | 60 t/s |
| 356 | Asyncflow V2 Async | 40.0 | 60 t/s |
| 357 | Claude Sonnet 5 High Anthropic | 40.0 | 80 t/s |
| 358 | Claude Sonnet 5 Low Anthropic | 40.0 | 80 t/s |
| 359 | Claude Sonnet 5 Medium Anthropic | 40.0 | 80 t/s |
| 360 | Claude Sonnet 5 Xhigh Anthropic | 40.0 | 80 t/s |
| 361 | Cogito V2 1 Reasoning Cogito | 40.0 | 40 t/s |
| 362 | Falcon 2 TII · Open | 40.0 | 60 t/s |
| 363 | Gladia Solaria Upstage | 40.0 | 60 t/s |
| 364 | Gpt 5 4 Pro OpenAI | 40.0 | 90 t/s |
| 365 | Gpt Realtime 2 OpenAI | 40.0 | 90 t/s |
| 366 | Grok Stt xAI | 40.0 | 80 t/s |
| 367 | Hidream O1 Image OpenAI | 40.0 | 60 t/s |
| 368 | Mistral Mistral · Open | 40.0 | 100 t/s |
| 369 | Nvidia NVIDIA | 40.0 | 150 t/s |
| 370 | 40.0 | 150 t/s | |
| 371 | Ocateve 2 Ocat | 40.0 | 60 t/s |
| 372 | Qwen Image Alibaba · Open | 40.0 | 85 t/s |
| 373 | Qwen3 Asr Alibaba · Open | 40.0 | 85 t/s |
| 374 | Solar Pro 3 Upstage | 34.0 | 40 t/s |
| 375 | GPT-oss-20B OpenAI · Open | 32.0 | 90 t/s |
| 376 | 31.0 | 40 t/s | |
| 377 | GPT-5.3 Codex OpenAI | 0.0 | 118 t/s |
Podium Score is a weighted blend of five leaderboards. See Methodology for details.