← Back to Test Cases
According to EU GMP Annex 11, what specific requirements apply to data storage and archiving for computerised systems?
GMP KnowledgeEu Annex11hardretired
This question has been retired from scoring. Every model answered it correctly on every attempt, so it no longer separates them. It stays published for reference, and historical results against it remain in the database.
Cross-Model Comparison
| Model | Score | Latency | Tokens In | Tokens Out |
|---|---|---|---|---|
| GPT-5.4 | 100.0% | 2.6s | 119 | 149 |
| GPT-5.4 mini | 100.0% | 1.1s | 119 | 88 |
| GPT-5.4 nano | 100.0% | 1.5s | 119 | 205 |
| Claude Haiku 4.5 | 100.0% | 3.7s | 127 | 301 |
| Claude Sonnet 4.6 | 100.0% | 9.7s | 127 | 382 |
| Llama 4 Scout | 100.0% | 1.2s | 124 | 437 |
| Claude Opus 4.6 | 100.0% | 11.6s | 127 | 463 |
| Llama 4 Maverick | 100.0% | 3.6s | 125 | 463 |
| Llama 3.3 70B Instruct | 100.0% | 6.4s | 125 | 160 |
| Mistral Small 2603 | 100.0% | 3.9s | 132 | 500 |
| DeepSeek V4 Pro | 100.0% | 9.4s | 118 | 365 |
| Qwen3.5-397B-A17B | 100.0% | 11.2s | 126 | 1,201 |
| DeepSeek-V3.2 | 100.0% | 56.2s | 118 | 495 |
| DeepSeek-R1 | 100.0% | 12.0s | 124 | 402 |
| DeepSeek-R1-Distill-Qwen-32B | 100.0% | 45.8s | 118 | 485 |
| Mistral Large 3 675B | 100.0% | 6.7s | 120 | 406 |
| Qwen3.5-35B-A3B | 100.0% | 7.5s | 126 | 1,162 |
| Gemini 3.1 Pro | 100.0% | 11.7s | 114 | 1,081 |
| Gemini 3 Flash | 100.0% | 3.3s | 114 | 315 |
| Gemini 3.1 Flash-Lite | 100.0% | 2.3s | 115 | 246 |
| Qwen3.6 27B | 100.0% | 18.4s | 126 | 1,267 |
| Qwen3.6 35B A3B | 100.0% | 10.2s | 126 | 1,808 |
| DeepSeek-V3.2 | 100.0% | 9.4s | 118 | 187 |
| DeepSeek V4 Flash | 100.0% | 3.1s | 118 | 421 |
| Gemma 4 26B A4B IT | 100.0% | 3.1s | 127 | 325 |
| Gemma 4 31B IT | 100.0% | 5.5s | 127 | 248 |
| MiniMax M2.7 | 100.0% | 7.7s | 155 | 351 |
| Claude Opus 4.8 | 100.0% | 8.4s | 167 | 659 |
| GLM-5.2 | 100.0% | 28.7s | 126 | 1,052 |
Tags
data_storagearchivingdata_integrityretention