← Back to Test Cases
Write Equipment Cleaning SOP
Task CompletionSopmedium
Cross-Model Comparison
| Model | Score | Latency | Tokens In | Tokens Out |
|---|---|---|---|---|
| MiniMax M3 | N/A | N/A | - | - |
| Muse Glimmer 30B | N/A | N/A | - | - |
| Hy4 Preview | N/A | N/A | - | - |
| Hy4 Preview | N/A | N/A | - | - |
| Hy4 Preview | N/A | N/A | - | - |
| GLM-5.3 | N/A | N/A | - | - |
| GLM-5.3 | N/A | N/A | - | - |
| GLM-5.3 | N/A | N/A | - | - |
| MiniMax M3 | N/A | N/A | - | - |
| MiniMax M3 | N/A | N/A | - | - |
| Qwen3.8-Flash | N/A | N/A | - | - |
| Qwen3.8-Flash | N/A | N/A | - | - |
| Qwen3.8-Flash | N/A | N/A | - | - |
| Qwen3.8 Max (2.4T-A95B) | N/A | N/A | - | - |
| Qwen3.8 Max (2.4T-A95B) | N/A | N/A | - | - |
| Qwen3.8 Max (2.4T-A95B) | N/A | N/A | - | - |
| Muse Glimmer 30B | N/A | N/A | - | - |
| Muse Glimmer 30B | N/A | N/A | - | - |
| Claude Opus 4.6 | 93.7% | 232.0s | 194 | 12,123 |
| Claude Opus 4.8 | 92.3% | 73.7s | 297 | 6,004 |
| Claude Opus 5 | 90.8% | 259.8s | 297 | 19,349 |
| Claude Haiku 4.5 | 90.6% | 102.9s | 193 | 13,010 |
| DeepSeek V4 Pro | 90.4% | 49.9s | 168 | 2,938 |
| GPT-5.4 | 90.3% | 64.4s | 175 | 3,697 |
| Claude Sonnet 4.6 | 90.2% | 269.8s | 194 | 16,000 |
| Qwen3.5-397B-A17B | 90.0% | 55.2s | 190 | 4,491 |
| Kimi K3 | 89.7% | 389.0s | 265 | 11,092 |
| Claude Opus 5 | 89.6% | 239.4s | 297 | 18,381 |
| Claude Opus 5 | 89.6% | 287.0s | 297 | 21,260 |
| Claude Sonnet 5 | 89.6% | 60.4s | 297 | 5,591 |
| Qwen3.8 Max (2.4T-A95B) | 89.1% | 97.2s | 228 | 18,809 |
| Qwen3.5-35B-A3B | 88.8% | 30.9s | 190 | 3,884 |
| GLM-5.3-Flash | 88.8% | 269.5s | 180 | 10,497 |
| DeepSeek V4 Pro | 88.5% | 212.4s | 247 | 9,949 |
| GLM-5.3 | 88.3% | 88.8s | 180 | 10,777 |
| Qwen3.8 Max (2.4T-A95B) | 88.0% | 639.7s | 228 | 24,982 |
| GLM-5.3 | 88.0% | 25.7s | 181 | 3,110 |
| Kimi K3 | 87.8% | 277.4s | 265 | 7,957 |
| Claude Sonnet 5 | 87.5% | 75.3s | 297 | 6,904 |
| GPT-5.6 Sol | 87.2% | 107.2s | 175 | 7,233 |
| Qwen3.8-Flash | 87.1% | 239.9s | 228 | 24,082 |
| DeepSeek V4 Pro | 86.8% | 26.3s | 169 | 2,906 |
| GLM-5.3-Flash | 86.7% | 193.8s | 180 | 7,407 |
| Qwen3.6 27B | 86.6% | 68.8s | 190 | 4,272 |
| GLM-5.2 | 86.6% | 31.0s | 180 | 2,437 |
| DeepSeek-R1 | 86.6% | 75.7s | 168 | 2,821 |
| GLM-5.3 | 86.5% | 113.0s | 180 | 10,354 |
| GLM-5.3-Flash | 86.5% | 224.2s | 180 | 7,788 |
| GPT-5.6 Sol | 86.4% | 111.0s | 175 | 7,707 |
| Mistral Large 3 675B | 86.4% | 33.1s | 182 | 1,828 |
| DeepSeek V4 Flash | 86.4% | 25.7s | 168 | 2,384 |
| Qwen3.6 35B A3B | 86.1% | 30.3s | 190 | 4,210 |
| MiniMax M2.7 | 86.1% | 101.3s | 178 | 5,913 |
| GPT-5.6 Sol | 85.8% | 88.3s | 175 | 5,939 |
| MiniMax M3 | 85.7% | 36.0s | 332 | 7,944 |
| Qwen3.5-397B-A17B | 85.7% | 62.1s | 190 | 4,371 |
| Kimi K3 | 85.4% | 369.1s | 265 | 11,067 |
| GPT-5.6 Terra | 85.2% | 96.6s | 175 | 6,047 |
| Qwen3.8 27B | 84.7% | 284.6s | 228 | 19,079 |
| MiniMax M2.7 | 84.7% | 68.2s | 176 | 5,139 |
| GPT-5.6 Terra | 84.5% | 100.1s | 175 | 6,604 |
| GPT-5.6 Terra | 84.5% | 91.5s | 175 | 5,990 |
| Qwen3.8-Flash | 84.3% | 136.0s | 228 | 14,935 |
| Qwen3.8-Flash | 83.7% | 234.2s | 228 | 26,048 |
| Claude Sonnet 5 | 83.6% | 42.6s | 297 | 4,028 |
| Qwen3.5-35B-A3B | 83.4% | 24.0s | 190 | 3,554 |
| DeepSeek V4 Flash | 83.4% | 84.8s | 168 | 2,848 |
| Gemini 3.1 Pro | 83.1% | 44.2s | 181 | 3,430 |
| Mistral Small 2603 | 83.1% | 10.9s | 194 | 1,806 |
| Qwen3.6 27B | 82.8% | 132.1s | 190 | 4,590 |
| Hy4 Preview | 82.8% | 226.5s | 189 | 17,389 |
| GPT-5.4 nano | 82.5% | 14.3s | 175 | 2,911 |
| MiniMax M2.7 | 82.3% | 169.9s | 178 | 8,283 |
| MiniMax M3 | 82.2% | 37.2s | 280 | 4,836 |
| Qwen3.8 27B | 82.2% | 98.2s | 296 | 14,454 |
| MiniMax M3 | 82.0% | 35.8s | 280 | 4,260 |
| DeepSeek-V3.2 | 81.6% | 79.4s | 168 | 2,331 |
| MiniMax M2.7 | 81.3% | 82.6s | 176 | 6,458 |
| DeepSeek V4 Flash | 81.0% | 34.7s | 168 | 2,418 |
| GPT-5.6 Luna | 81.0% | 50.9s | 175 | 5,012 |
| DeepSeek-V3.2 | 80.8% | 159.9s | 168 | 2,502 |
| DeepSeek-V3.2 | 80.8% | 24.6s | 168 | 2,321 |
| Qwen3.6 27B | 80.5% | 114.9s | 190 | 3,961 |
| Qwen3.8 27B | 80.3% | 170.8s | 296 | 14,358 |
| Qwen3.5-35B-A3B | 80.2% | 32.9s | 190 | 3,787 |
| DeepSeek V4 Flash | 80.2% | 43.6s | 168 | 2,548 |
| GPT-5.6 Luna | 80.1% | 50.2s | 175 | 4,896 |
| Qwen3.8 Max (2.4T-A95B) | 79.9% | 742.4s | 228 | 31,072 |
| Qwen3.6 35B A3B | 79.8% | 15.9s | 190 | 4,250 |
| Qwen3.5-397B-A17B | 79.7% | 41.5s | 190 | 3,658 |
| Muse Glimmer 30B | 79.7% | 33.6s | 193 | 2,213 |
| Mistral Large 3 675B | 79.4% | 25.8s | 182 | 2,033 |
| Hy4 Preview | 79.2% | 238.7s | 189 | 18,299 |
| Qwen3.6 27B | 78.5% | 68.7s | 190 | 4,284 |
| Qwen3.5-35B-A3B | 78.5% | 22.7s | 190 | 3,693 |
| Mistral Large 3 675B | 78.3% | 27.0s | 182 | 2,204 |
| Hy4 Preview | 78.3% | 213.8s | 189 | 16,014 |
| GLM-5.2 | 78.2% | 67.2s | 180 | 3,448 |
| Qwen3.6 35B A3B | 78.1% | 35.3s | 190 | 4,733 |
| GLM-5.2 | 77.9% | 70.1s | 180 | 3,496 |
| GPT-5.4 mini | 77.5% | 14.4s | 175 | 2,423 |
| Mistral Small 2603 | 77.0% | 14.6s | 194 | 2,017 |
| DeepSeek-V3.2 | 76.7% | 49.1s | 168 | 2,309 |
| DeepSeek-R1 | 76.5% | 154.8s | 169 | 3,099 |
| Claude Haiku 4.5 | 76.4% | 106.6s | 193 | 13,899 |
| Gemma 4 26B A4B IT | 76.2% | 16.1s | 197 | 1,512 |
| Claude Haiku 4.5 | 75.8% | 77.9s | 193 | 9,313 |
| DeepSeek-V3.2 | 75.8% | 48.7s | 168 | 2,333 |
| Qwen3.5-397B-A17B | 75.7% | 46.8s | 190 | 4,286 |
| DeepSeek V4 Pro | 75.2% | 79.4s | 168 | 3,383 |
| Mistral Small 2603 | 75.0% | 15.1s | 194 | 2,183 |
| Gemma 4 26B A4B IT | 75.0% | 16.2s | 197 | 1,472 |
| GPT-5.6 Luna | 75.0% | 52.6s | 175 | 5,143 |
| Mistral Small 2603 | 75.0% | 14.8s | 194 | 2,037 |
| Claude Haiku 4.5 | 74.7% | 68.4s | 193 | 8,242 |
| Qwen3.6 35B A3B | 74.6% | 22.3s | 190 | 4,479 |
| Muse Glimmer 30B | 73.6% | 33.2s | 193 | 2,165 |
| Gemma 4 31B IT | 73.5% | 41.5s | 196 | 1,444 |
| Muse Glimmer 30B | 73.3% | 34.2s | 193 | 2,203 |
| Mistral Large 3 675B | 72.8% | 24.1s | 182 | 2,002 |
| DeepSeek-R1 | 72.7% | 143.9s | 169 | 2,799 |
| GLM-5.2 | 71.5% | 94.8s | 180 | 3,507 |
| Gemma 4 31B IT | 71.0% | 7.2s | 196 | 1,386 |
| Gemini 3.1 Flash-Lite | 70.2% | 6.7s | 181 | 986 |
| Llama 4 Maverick | 68.8% | 33.5s | 180 | 849 |
| Gemini 3 Flash | 68.2% | 9.2s | 179 | 1,328 |
| Llama 3.3 70B Instruct | 65.8% | 28.8s | 183 | 865 |
| DeepSeek-R1 | 65.8% | 75.8s | 167 | 1,285 |
| Llama 4 Maverick | 65.7% | 37.6s | 180 | 908 |
| Gemma 4 31B IT | 65.3% | 8.3s | 196 | 1,390 |
| Gemma 4 26B A4B IT | 64.8% | 20.3s | 196 | 1,403 |
| Gemma 4 26B A4B IT | 63.4% | 32.6s | 212 | 1,361 |
| Gemma 4 31B IT | 62.7% | 54.8s | 212 | 1,269 |
| Llama 4 Scout | 61.8% | 7.8s | 180 | 697 |
| Llama 3.3 70B Instruct | 59.8% | 24.8s | 202 | 876 |
| Llama 3.3 70B Instruct | 58.0% | 15.7s | 183 | 830 |
| Llama 4 Maverick | 56.6% | 32.5s | 180 | 722 |
| Llama 4 Maverick | 56.4% | 18.2s | 180 | 854 |
| Llama 4 Scout | 54.0% | 17.1s | 180 | 831 |
| Llama 3.3 70B Instruct | 52.8% | 25.5s | 202 | 809 |
| Llama 4 Scout | 52.3% | 16.1s | 180 | 783 |
| Llama 4 Scout | 52.3% | 25.0s | 180 | 722 |
| DeepSeek-R1-Distill-Qwen-32B | 50.2% | 57.8s | 176 | 1,193 |
Tags
sopcleaningbioreactor