Models · GPT
GPT-5.6 Luna
Asked for a few hundred random integers, GPT-5.6 Luna does not pick evenly. This page is the shape of that preference, built from 36 recorded answers containing 12775 numbers, and it is what the identify tool compares a pasted reply against.
- Bank id
gpt-5.6-luna- Family
- GPT
- Collected through
- API
- Status
- Inherited from ModelTrace; serving identity not independently verified
- Distinct values used
- 355 of 355
- Prompt environments
- 12 of 12
Distribution
Every bar is one value from 1 to 355; its height is how many times the model chose it across all recorded answers. A perfectly random picker would give a flat line. Nothing in the bank does.
Favourite numbers
| Number | Times chosen | Share |
|---|---|---|
| 346 | 68 | 0.53% |
| 347 | 62 | 0.49% |
| 306 | 61 | 0.48% |
| 319 | 61 | 0.48% |
| 278 | 56 | 0.44% |
| 279 | 55 | 0.43% |
| 289 | 55 | 0.43% |
| 326 | 55 | 0.43% |
| 333 | 55 | 0.43% |
| 246 | 52 | 0.41% |
| 340 | 52 | 0.41% |
| 316 | 51 | 0.40% |
How reliably it is recognised
In held-out testing across all twelve prompt environments, 12 of 12 three-answer runs named this model, 0 narrowed it to its family, 0 gave no answer, and 0 named a different model.
With this model removed from the bank and its own answers scored against the rest, 0 of 12 runs still confidently named another model, 10 stopped at the family, and 2 gave no answer. Its answers landed most often on GPT-5.5 (6 of 12). That is the number to keep in mind if you are testing a version that is not listed here.
Both measurements use the open-set verdict the site applies, not the bank's raw ranking. The methodology page explains how they were made and what they leave out.