← Back to Test Cases
annex11-001
GMP KnowledgeEu Annex11easy
Cross-Model Comparison
| Model | Score | Latency | Tokens In | Tokens Out |
|---|---|---|---|---|
| GLM-5.3 | N/A | N/A | - | - |
| MiniMax M3 | N/A | N/A | - | - |
| Hy4 Preview | N/A | N/A | - | - |
| Muse Glimmer 30B | N/A | N/A | - | - |
| Qwen3.8 Max (2.4T-A95B) | N/A | N/A | - | - |
| Qwen3.8-Flash | N/A | N/A | - | - |
| Claude Opus 4.6 | 100.0% | 10.1s | 97 | 395 |
| Llama 4 Maverick | 100.0% | 10.4s | 96 | 396 |
| Llama 3.3 70B Instruct | 100.0% | 8.6s | 97 | 186 |
| Mistral Small 2603 | 100.0% | 1.2s | 104 | 207 |
| Qwen3.5-397B-A17B | 100.0% | 6.8s | 100 | 835 |
| DeepSeek-V3.2 | 100.0% | 10.6s | 89 | 129 |
| DeepSeek V4 Flash | 100.0% | 2.2s | 228 | 79 |
| DeepSeek-R1 | 100.0% | 15.1s | 95 | 558 |
| DeepSeek-R1-Distill-Qwen-32B | 100.0% | 30.3s | 92 | 478 |
| Mistral Large 3 675B | 100.0% | 5.2s | 92 | 360 |
| Gemini 3.1 Pro | 100.0% | 14.5s | 88 | 865 |
| Gemini 3 Flash | 100.0% | 8.1s | 88 | 268 |
| Gemini 3.1 Flash-Lite | 100.0% | 1.1s | 88 | 100 |
| Qwen3.6 27B | 100.0% | 9.8s | 100 | 896 |
| Qwen3.6 35B A3B | 100.0% | 5.6s | 100 | 1,006 |
| DeepSeek V4 Pro | 100.0% | 10.4s | 89 | 500 |
| DeepSeek-V3.2 | 100.0% | 3.9s | 95 | 89 |
| DeepSeek V4 Flash | 100.0% | 1.9s | 89 | 239 |
| Gemma 4 26B A4B IT | 100.0% | 4.3s | 101 | 230 |
| Gemma 4 31B IT | 100.0% | 4.3s | 101 | 231 |
| MiniMax M2.7 | 100.0% | 2.7s | 126 | 385 |
| Llama 4 Maverick | 100.0% | 5.9s | 153 | 301 |
| Claude Opus 4.8 | 100.0% | 7.0s | 129 | 476 |
| GLM-5.2 | 100.0% | 4.3s | 98 | 298 |
| Llama 4 Scout | 100.0% | 8.6s | 153 | 313 |
| Llama 3.3 70B Instruct | 100.0% | 19.7s | 156 | 308 |
| Mistral Small 2603 | 100.0% | 2.4s | 167 | 327 |
| Mistral Large 3 675B | 100.0% | 3.3s | 155 | 240 |
| Qwen3.5-35B-A3B | 100.0% | 7.0s | 162 | 1,064 |
| Qwen3.6 27B | 100.0% | 18.7s | 162 | 615 |
| Hy4 Preview | 100.0% | 16.8s | 166 | 1,080 |
| Qwen3.6 35B A3B | 100.0% | 46.7s | 162 | 1,147 |
| Qwen3.5-397B-A17B | 100.0% | 13.1s | 162 | 937 |
| DeepSeek V4 Pro | 100.0% | 5.0s | 149 | 300 |
| DeepSeek-V3.2 | 100.0% | 3.1s | 149 | 64 |
| Gemma 4 26B A4B IT | 100.0% | 4.0s | 163 | 248 |
| DeepSeek-R1 | 100.0% | 75.1s | 148 | 1,052 |
| Gemma 4 26B A4B IT | 100.0% | 4.8s | 163 | 215 |
| Gemma 4 31B IT | 100.0% | 6.3s | 163 | 179 |
| GLM-5.2 | 100.0% | 4.6s | 158 | 200 |
| MiniMax M2.7 | 100.0% | 3.5s | 184 | 260 |
| Muse Glimmer 30B | 100.0% | 3.2s | 199 | 265 |
| Qwen3.8 27B | 100.0% | 3.5s | 204 | 121 |
| Kimi K3 | 100.0% | 14.0s | 231 | 422 |
| GLM-5.3-Flash | 100.0% | 4.3s | 158 | 118 |
| GPT-5.6 Luna | 100.0% | 1.3s | 148 | 71 |
| GPT-5.4 mini | 100.0% | 928ms | 90 | 61 |
| GPT-5.4 nano | 100.0% | 834ms | 90 | 70 |
| Claude Haiku 4.5 | 100.0% | 3.5s | 97 | 297 |
| Claude Sonnet 4.6 | 100.0% | 9.4s | 97 | 376 |
| Llama 4 Scout | 100.0% | 1.2s | 95 | 322 |
| GPT-5.4 | 100.0% | 1.1s | 90 | 52 |
| Qwen3.8 Max (2.4T-A95B) | 100.0% | 1.1s | 204 | 68 |
| Qwen3.8-Flash | 100.0% | 4.8s | 213 | 105 |
| MiniMax M3 | 100.0% | 4.1s | 319 | 281 |
| GPT-5.6 Sol | 100.0% | 1.5s | 148 | 42 |
| GLM-5.3 | 100.0% | 4.6s | 158 | 351 |
| Claude Opus 5 | 100.0% | 5.0s | 217 | 275 |
| Claude Sonnet 5 | 100.0% | 1.2s | 217 | 9 |
| Claude Haiku 4.5 | 100.0% | 9.4s | 161 | 318 |
| GPT-5.6 Terra | 100.0% | 2.0s | 148 | 47 |
| Qwen3.5-35B-A3B | 0.0% | 7.0s | 100 | 1,190 |
Tags
validationrisk_assessmentcomputerised_systems