| 1 | Z.ai: GLM 5.2 | z-ai | 83.2% |
| 2 | DeepSeek: DeepSeek V4 Flash 0423 | deepseek | 81.8% |
| 3 | MoonshotAI: Kimi K3 | moonshotai | 80.3% |
| 4 | DeepSeek: DeepSeek V4 Pro 0423 | deepseek | 80.3% |
| 5 | NVIDIA: Nemotron 3 Ultra | nvidia | 75.2% |
| 6 | Xiaomi: MiMo-V2.5-Pro | xiaomi | 71.5% |
| 7 | MoonshotAI: Kimi K2.6 | moonshotai | 66.9% |
| 8 | MoonshotAI: Kimi K2.7 Code | moonshotai | 66.9% |
| 9 | Xiaomi: MiMo-V2.5 | xiaomi | 66.5% |
| 10 | Qwen: Qwen3 VL 30B A3B Instruct | qwen | 66.1% |
| 11 | Qwen: Qwen3 VL 32B Instruct | qwen | 66.1% |
| 12 | Nex AGI: Nex-N2-Pro (free) | nex-agi | 65.7% |
| 13 | Z.ai: GLM 4.6V | z-ai | 65.7% |
| 14 | Qwen: Qwen3 VL 235B A22B Instruct | qwen | 64.4% |
| 15 | Qwen: Qwen3 VL 30B A3B Thinking | qwen | 64.0% |
| 16 | Google: Gemma 4 31B | google | 63.6% |
| 17 | Z.ai: GLM 5.1 | z-ai | 63.5% |
| 18 | Baidu: ERNIE 4.5 VL 424B A47B | baidu | 63.2% |
| 19 | Qwen: Qwen3 VL 8B Thinking | qwen | 63.2% |
| 20 | Mistral: Mistral Small 4 | mistralai | 57.7% |
| 21 | Qwen: Qwen3 VL 8B Instruct | qwen | 56.9% |
| 22 | Mistral: Ministral 3 14B 2512 | mistralai | 56.5% |
| 23 | Google: Gemma 4 26B A4B | google | 55.2% |
| 24 | Tencent: Hy3 preview | tencent | 54.0% |
| 25 | Mistral: Ministral 3 8B 2512 | mistralai | 54.0% |
| 26 | Mistral: Mistral Small 3.2 24B | mistralai | 53.6% |
| 27 | Meta: Llama 4 Maverick | meta-llama | 52.7% |
| 28 | Google: Gemma 3 27B | google | 50.2% |
| 29 | Qwen: Qwen2.5 VL 72B Instruct | qwen | 50.2% |
| 30 | MiniMax: MiniMax-01 | minimax | 49.4% |
| 31 | Qwen: Qwen3.5-27B | qwen | 49.4% |
| 32 | Google: Gemma 3 12B | google | 49.0% |
| 33 | Meta: Llama 4 Scout | meta-llama | 48.5% |
| 34 | Qwen: Qwen3.6 35B A3B | qwen | 47.7% |
| 35 | Google: Gemma 3 4B | google | 47.3% |
| 36 | Mistral: Ministral 3 3B 2512 | mistralai | 45.6% |
| 37 | NVIDIA: Nemotron Nano 12B 2 VL (free) | nvidia | 42.3% |
| 38 | Arcee AI: Trinity Large Thinking | arcee-ai | 41.6% |
| 39 | Qwen: Qwen3.5-9B | qwen | 41.4% |
| 40 | Qwen: Qwen3.5-35B-A3B | qwen | 39.3% |
| 41 | Meta: Llama 3.2 11B Vision Instruct | meta-llama | 35.6% |
| 42 | AI21: Jamba Large 1.7 | ai21 | — |
| 43 | AionLabs: Aion-1.0-Mini | aion-labs | — |
| 44 | AllenAI: Olmo 3 32B Think | allenai | — |
| 45 | Arcee AI: Trinity Mini | arcee-ai | — |
| 46 | ByteDance: UI-TARS 7B | bytedance | — |
| 47 | Cohere: Command A | cohere | — |
| 48 | Cohere: North Mini Code (free) | cohere | — |
| 49 | DeepSeek: DeepSeek V3 | deepseek | — |
| 50 | DeepSeek: DeepSeek V3 0324 | deepseek | — |
| 51 | DeepSeek: DeepSeek V3.1 | deepseek | — |
| 52 | DeepSeek: DeepSeek V3.1 Terminus | deepseek | — |
| 53 | DeepSeek: DeepSeek V3.2 | deepseek | — |
| 54 | DeepSeek: DeepSeek V3.2 Exp | deepseek | — |
| 55 | DeepSeek: DeepSeek V4 Flash 0731 | deepseek | — |
| 56 | DeepSeek: DeepSeek V4 Flash Vision Exp | deepseek | — |
| 57 | DeepSeek: DeepSeek V4 Pro 0813 | deepseek | — |
| 58 | DeepSeek: DeepSeek V4.1 Flash | deepseek | — |
| 59 | DeepSeek: DeepSeek V4.1 Flash (batch) | deepseek | — |
| 60 | DeepSeek: R1 | deepseek | — |
| 61 | DeepSeek: R1 0528 | deepseek | — |
| 62 | DeepSeek: R1 Distill Llama 70B | deepseek | — |
| 63 | DeepSeek: R1 Distill Qwen 32B | deepseek | — |
| 64 | EssentialAI: Rnj 1 Instruct | essentialai | — |
| 65 | Google: Gemma 2 27B | google | — |
| 66 | Google: Gemma 3n 4B | google | — |
| 67 | Google: Gemma 4 26B A4B (free) | google | — |
| 68 | Google: Gemma 4 31B (free) | google | — |
| 69 | IBM: Granite 4.0 Micro | ibm-granite | — |
| 70 | IBM: Granite 4.1 8B | ibm-granite | — |
| 71 | IBM: Granite 4.2 8B | ibm-granite | — |
| 72 | inclusionAI: Ling 3.0 Flash | inclusionai | — |
| 73 | inclusionAI: Ling 3.0 Flash VL | inclusionai | — |
| 74 | Inference.net: Schematron V2 Small | inference-net | — |
| 75 | Inference.net: Schematron V2 Turbo | inference-net | — |
| 76 | LiquidAI: LFM2-24B-A2B | liquid | — |
| 77 | LiquidAI: LFM2.5-1.2B-Instruct (free) | liquid | — |
| 78 | LiquidAI: LFM2.5-1.2B-Thinking (free) | liquid | — |
| 79 | LiquidAI: LFM2.5-2.6B (free) | liquid | — |
| 80 | LiquidAI: LFM2.5-24B-A2B | liquid | — |
| 81 | Llama Guard 3 8B | meta-llama | — |
| 82 | Magnum v4 72B | anthracite-org | — |
| 83 | Meituan: LongCat 2.0 | meituan | — |
| 84 | Meta: Llama 3 70B Instruct | meta-llama | — |
| 85 | Meta: Llama 3 8B Instruct | meta-llama | — |
| 86 | Meta: Llama 3.1 70B Instruct | meta-llama | — |
| 87 | Meta: Llama 3.1 8B Instruct | meta-llama | — |
| 88 | Meta: Llama 3.2 1B Instruct | meta-llama | — |
| 89 | Meta: Llama 3.2 3B Instruct | meta-llama | — |
| 90 | Meta: Llama 3.2 3B Instruct (free) | meta-llama | — |
| 91 | Meta: Llama 3.3 70B Instruct | meta-llama | — |
| 92 | Meta: Llama 3.3 70B Instruct (free) | meta-llama | — |
| 93 | Meta: Llama Guard 4 12B | meta-llama | — |
| 94 | Meta: Muse Glimmer 30B | meta | — |
| 95 | Microsoft: Phi 4 | microsoft | — |
| 96 | Microsoft: Phi 4 Mini Instruct | microsoft | — |
| 97 | MiniMax: MiniMax M2 | minimax | — |
| 98 | MiniMax: MiniMax M2.1 | minimax | — |
| 99 | MiniMax: MiniMax M2.5 | minimax | — |
| 100 | MiniMax: MiniMax M2.7 | minimax | — |
| 101 | Mistral: Devstral 2 2512 | mistralai | — |
| 102 | Mistral: Ministral 3 8B 2512 (batch) | mistralai | — |
| 103 | Mistral: Mistral Nemo | mistralai | — |
| 104 | Mistral: Mistral Small 3 | mistralai | — |
| 105 | Mistral: Mistral Small 3.1 24B | mistralai | — |
| 106 | Mistral: Mistral Small 4 (batch) | mistralai | — |
| 107 | Mistral: Mixtral 8x22B Instruct | mistralai | — |
| 108 | Mistral: Voxtral Small 24B 2507 | mistralai | — |
| 109 | MoonshotAI: Kimi K2 0711 | moonshotai | — |
| 110 | MoonshotAI: Kimi K2 0905 | moonshotai | — |
| 111 | MoonshotAI: Kimi K2 Thinking | moonshotai | — |
| 112 | MoonshotAI: Kimi K2.5 | moonshotai | — |
| 113 | MoonshotAI: Kimi K2.6 (free) | moonshotai | — |
| 114 | MoonshotAI: Kimi K3 (batch) | moonshotai | — |
| 115 | MythoMax 13B | gryphe | — |
| 116 | Nex AGI: DeepSeek V3.1 Nex N1 | nex-agi | — |
| 117 | Nex AGI: Nex-N2-Pro | nex-agi | — |
| 118 | Nex AGI: Nex-N2.5-Mini | nex-agi | — |
| 119 | Nex AGI: Nex-N2.5-Pro | nex-agi | — |
| 120 | Nous: Hermes 3 405B Instruct | nousresearch | — |
| 121 | Nous: Hermes 3 405B Instruct (free) | nousresearch | — |
| 122 | Nous: Hermes 3 70B Instruct | nousresearch | — |
| 123 | Nous: Hermes 4 405B | nousresearch | — |
| 124 | Nous: Hermes 4 70B | nousresearch | — |
| 125 | NVIDIA: Llama 3.3 Nemotron Super 49B V1.5 | nvidia | — |
| 126 | NVIDIA: Nemotron 3 Nano 30B A3B | nvidia | — |
| 127 | NVIDIA: Nemotron 3 Nano 30B A3B (free) | nvidia | — |
| 128 | NVIDIA: Nemotron 3 Super | nvidia | — |
| 129 | NVIDIA: Nemotron 3 Super (free) | nvidia | — |
| 130 | NVIDIA: Nemotron 3 Ultra (free) | nvidia | — |
| 131 | NVIDIA: Nemotron 3.5 Content Safety | nvidia | — |
| 132 | NVIDIA: Nemotron 3.5 Content Safety (free) | nvidia | — |
| 133 | NVIDIA: Nemotron 3.5 Lightning | nvidia | — |
| 134 | NVIDIA: Nemotron 3.5 Lightning (free) | nvidia | — |
| 135 | NVIDIA: Nemotron Nano 9B V2 | nvidia | — |
| 136 | NVIDIA: Nemotron Nano 9B V2 (free) | nvidia | — |
| 137 | OpenAI: gpt-oss-120b | openai | — |
| 138 | OpenAI: gpt-oss-120b (batch) | openai | — |
| 139 | OpenAI: gpt-oss-120b (free) | openai | — |
| 140 | OpenAI: gpt-oss-20b | openai | — |
| 141 | OpenAI: gpt-oss-20b (batch) | openai | — |
| 142 | OpenAI: gpt-oss-20b (free) | openai | — |
| 143 | OpenAI: gpt-oss-safeguard-20b | openai | — |
| 144 | Poolside: Laguna M.1 | poolside | — |
| 145 | Poolside: Laguna S 2.1 | poolside | — |
| 146 | Poolside: Laguna S 2.1 (free) | poolside | — |
| 147 | Poolside: Laguna XS 2.1 | poolside | — |
| 148 | Poolside: Laguna XS 2.1 (free) | poolside | — |
| 149 | Poolside: Laguna XS.2 | poolside | — |
| 150 | Poolside: Laguna XS.2 (free) | poolside | — |
| 151 | Prime Intellect: INTELLECT-3 | prime-intellect | — |
| 152 | PrismML: Ternary Bonsai 2 27B | prism-ml | — |
| 153 | Qwen: Qwen2.5 7B Instruct | qwen | — |
| 154 | Qwen: Qwen3 14B | qwen | — |
| 155 | Qwen: Qwen3 235B A22B | qwen | — |
| 156 | Qwen: Qwen3 235B A22B Instruct 2507 | qwen | — |
| 157 | Qwen: Qwen3 235B A22B Thinking 2507 | qwen | — |
| 158 | Qwen: Qwen3 30B A3B | qwen | — |
| 159 | Qwen: Qwen3 30B A3B Instruct 2507 | qwen | — |
| 160 | Qwen: Qwen3 30B A3B Thinking 2507 | qwen | — |
| 161 | Qwen: Qwen3 32B | qwen | — |
| 162 | Qwen: Qwen3 8B | qwen | — |
| 163 | Qwen: Qwen3 Coder 30B A3B Instruct | qwen | — |
| 164 | Qwen: Qwen3 Coder 480B A35B | qwen | — |
| 165 | Qwen: Qwen3 Coder 480B A35B (free) | qwen | — |
| 166 | Qwen: Qwen3 Coder Next | qwen | — |
| 167 | Qwen: Qwen3 Next 80B A3B Instruct | qwen | — |
| 168 | Qwen: Qwen3 Next 80B A3B Instruct (free) | qwen | — |
| 169 | Qwen: Qwen3 Next 80B A3B Thinking | qwen | — |
| 170 | Qwen: Qwen3 VL 235B A22B Thinking | qwen | — |
| 171 | Qwen: Qwen3.5 397B A17B | qwen | — |
| 172 | Qwen: Qwen3.5-122B-A10B | qwen | — |
| 173 | Qwen: Qwen3.6 27B | qwen | — |
| 174 | Qwen: Qwen3.8 2.4T A95B | qwen | — |
| 175 | Qwen: Qwen3.8 27B | qwen | — |
| 176 | Qwen: Qwen3.8 Flash | qwen | — |
| 177 | Qwen2.5 72B Instruct | qwen | — |
| 178 | Qwen2.5 Coder 32B Instruct | qwen | — |
| 179 | Reka Edge | rekaai | — |
| 180 | Reka Flash 3 | rekaai | — |
| 181 | ReMM SLERP 13B | undi95 | — |
| 182 | Sao10K: Llama 3 8B Lunaris | sao10k | — |
| 183 | Sao10K: Llama 3.1 70B Hanami x1 | sao10k | — |
| 184 | Sao10K: Llama 3.1 Euryale 70B v2.2 | sao10k | — |
| 185 | Sao10K: Llama 3.3 Euryale 70B | sao10k | — |
| 186 | StepFun: Step 3.5 Flash | stepfun | — |
| 187 | Tencent: Hunyuan A13B Instruct | tencent | — |
| 188 | Tencent: Hy-MT2-1.8B | tencent | — |
| 189 | Tencent: Hy-MT2-30B-A3B | tencent | — |
| 190 | Tencent: Hy-MT2-7B | tencent | — |
| 191 | Tencent: Hy3 | tencent | — |
| 192 | Tencent: Hy4 preview | tencent | — |
| 193 | TheDrummer: Cydonia 24B V4.1 | thedrummer | — |
| 194 | TheDrummer: Rocinante 12B | thedrummer | — |
| 195 | TheDrummer: Skyfall 36B V2 | thedrummer | — |
| 196 | TheDrummer: UnslopNemo 12B | thedrummer | — |
| 197 | Thinking Machines: Inkling | thinkingmachines | — |
| 198 | Thinking Machines: Inkling (free) | thinkingmachines | — |
| 199 | Thinking Machines: Inkling Small | thinkingmachines | — |
| 200 | Thinking Machines: Inkling Small (free) | thinkingmachines | — |
| 201 | Venice: Uncensored | cognitivecomputations | — |
| 202 | Venice: Uncensored (free) | cognitivecomputations | — |
| 203 | WizardLM-2 8x22B | microsoft | — |
| 204 | Xiaomi: MiMo-V2-Flash | xiaomi | — |
| 205 | Xiaomi: MiMo-V2.6-Flash | xiaomi | — |
| 206 | Xiaomi: MiMo-V2.6-Pro | xiaomi | — |
| 207 | Z.ai: GLM 4.5 | z-ai | — |
| 208 | Z.ai: GLM 4.5 Air | z-ai | — |
| 209 | Z.ai: GLM 4.5 Air (free) | z-ai | — |
| 210 | Z.ai: GLM 4.5V | z-ai | — |
| 211 | Z.ai: GLM 4.6 | z-ai | — |
| 212 | Z.ai: GLM 4.7 | z-ai | — |
| 213 | Z.ai: GLM 4.7 Flash | z-ai | — |
| 214 | Z.ai: GLM 5 | z-ai | — |
| 215 | Z.ai: GLM 5.3 | z-ai | — |
| 216 | Z.ai: GLM 5.3 (batch) | z-ai | — |
| 217 | Z.ai: GLM 5.3 Flash | z-ai | — |
| 218 | Z.ai: GLM 5.3 Flash (batch) | z-ai | — |