← Back to Test Cases
gmp-001
GMP KnowledgeGeneral Gmpeasy
Cross-Model Comparison
| Model | Score | Latency | Tokens In | Tokens Out |
|---|---|---|---|---|
| Qwen3.8-Flash | N/A | N/A | - | - |
| 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 | - | - |
| Claude Sonnet 4.6 | 100.0% | 8.1s | 115 | 322 |
| Llama 4 Scout | 100.0% | 5.1s | 108 | 401 |
| Claude Opus 4.6 | 100.0% | 9.2s | 115 | 332 |
| Llama 4 Maverick | 100.0% | 6.4s | 109 | 406 |
| Llama 3.3 70B Instruct | 100.0% | 1.9s | 110 | 178 |
| Mistral Small 2603 | 100.0% | 1.7s | 119 | 279 |
| Qwen3.5-397B-A17B | 100.0% | 12.7s | 115 | 844 |
| DeepSeek-R1 | 100.0% | 12.1s | 107 | 406 |
| DeepSeek-R1-Distill-Qwen-32B | 100.0% | 52.1s | 107 | 615 |
| Mistral Large 3 675B | 100.0% | 3.4s | 107 | 199 |
| Gemini 3 Flash | 100.0% | 2.0s | 102 | 138 |
| DeepSeek V4 Pro | 100.0% | 6.6s | 101 | 354 |
| Gemini 3.1 Pro | 100.0% | 6.5s | 102 | 607 |
| Gemini 3.1 Flash-Lite | 100.0% | 1.4s | 102 | 143 |
| Qwen3.6 27B | 100.0% | 19.8s | 115 | 897 |
| DeepSeek V4 Flash | 100.0% | 3.2s | 161 | 156 |
| Qwen3.6 35B A3B | 100.0% | 5.1s | 115 | 1,066 |
| DeepSeek-V3.2 | 100.0% | 4.0s | 101 | 107 |
| DeepSeek V4 Flash | 100.0% | 4.2s | 101 | 405 |
| Gemma 4 26B A4B IT | 100.0% | 2.9s | 115 | 160 |
| Gemma 4 31B IT | 100.0% | 7.5s | 115 | 139 |
| Llama 4 Maverick | 100.0% | 11.0s | 166 | 330 |
| Claude Opus 4.8 | 100.0% | 5.4s | 150 | 327 |
| GLM-5.2 | 100.0% | 11.7s | 111 | 358 |
| Qwen3.8 27B | 100.0% | 8.2s | 219 | 277 |
| Llama 4 Scout | 100.0% | 8.0s | 166 | 225 |
| Llama 3.3 70B Instruct | 100.0% | 47.7s | 169 | 357 |
| Mistral Small 2603 | 100.0% | 2.8s | 182 | 341 |
| Qwen3.5-35B-A3B | 100.0% | 8.8s | 177 | 1,017 |
| Mistral Large 3 675B | 100.0% | 2.2s | 170 | 148 |
| Qwen3.6 27B | 100.0% | 30.5s | 177 | 1,112 |
| Qwen3.6 35B A3B | 100.0% | 6.9s | 177 | 1,409 |
| Hy4 Preview | 100.0% | 8.8s | 178 | 673 |
| Qwen3.5-397B-A17B | 100.0% | 6.9s | 177 | 568 |
| DeepSeek V4 Pro | 100.0% | 6.9s | 161 | 56 |
| DeepSeek-V3.2 | 100.0% | 6.2s | 167 | 133 |
| Claude Sonnet 5 | 100.0% | 1.3s | 238 | 9 |
| Gemma 4 26B A4B IT | 100.0% | 2.8s | 177 | 204 |
| DeepSeek-R1 | 100.0% | 59.8s | 160 | 927 |
| Gemma 4 26B A4B IT | 100.0% | 7.8s | 177 | 212 |
| Gemma 4 31B IT | 100.0% | 12.5s | 177 | 184 |
| MiniMax M2.7 | 100.0% | 5.6s | 198 | 419 |
| GLM-5.2 | 100.0% | 17.5s | 171 | 549 |
| Kimi K3 | 100.0% | 11.1s | 244 | 401 |
| Muse Glimmer 30B | 100.0% | 6.1s | 213 | 389 |
| GLM-5.3-Flash | 100.0% | 6.7s | 171 | 255 |
| GPT-5.6 Luna | 100.0% | 1.7s | 162 | 88 |
| GPT-5.4 mini | 100.0% | 736ms | 104 | 30 |
| MiniMax M2.7 | 100.0% | 2.5s | 139 | 310 |
| GPT-5.4 nano | 100.0% | 1.4s | 104 | 69 |
| Claude Haiku 4.5 | 100.0% | 2.4s | 115 | 273 |
| DeepSeek-V3.2 | 100.0% | 24.2s | 101 | 109 |
| GPT-5.4 | 100.0% | 1.6s | 104 | 67 |
| Qwen3.8 Max (2.4T-A95B) | 100.0% | 2.1s | 219 | 181 |
| Qwen3.8-Flash | 100.0% | 1.7s | 228 | 138 |
| MiniMax M3 | 100.0% | 1.8s | 333 | 324 |
| GLM-5.3 | 100.0% | 1.8s | 172 | 147 |
| Claude Opus 5 | 100.0% | 7.6s | 238 | 449 |
| Claude Haiku 4.5 | 100.0% | 2.5s | 179 | 232 |
| GPT-5.6 Sol | 100.0% | 2.1s | 162 | 47 |
| GPT-5.6 Terra | 100.0% | 1.8s | 162 | 68 |
| Qwen3.5-35B-A3B | 0.0% | 6.0s | 115 | 1,039 |
Tags
deviationsinvestigationbatch_disposition