Daily numbers
What each model did on each observed day. The same 100 questions are asked twice every day.
- Observed: questions answered on that day, out of those asked
- Same answer twice: questions whose answers were identical when asked twice on the same day, out of the observed questions
- Refused: boundary questions (where refusing is a reasonable choice) that the model refused, out of the observed boundary questions
- Wobble: how the two answers to the same question differed, out of the observed questions. Wording: same answer, different words. Answer: a different answer. Format: a different shape (code fences, lists, length). Refusal: it refused once and answered once
- Change: the day a change from the previous observed day was detected. See the model's page for what changed
Wobble alone is not a change. Newer models reject any temperature other than their default, so they wobble more every day.
All models on 2026-10-10
| Date | Observed | Same answer twice | Refused (boundary) | Wobble: wording | answer | format | refusal | Change |
|---|
| 2026-10-10 | 100/100 | 100/100 | 8/15 | 0/100 | 0/100 | 0/100 | 0/100 | |
| 2026-10-09 | 100/100 | 100/100 | 8/15 | 0/100 | 0/100 | 0/100 | 0/100 | changed |
| 2026-10-08 | 100/100 | 95/100 | 8/15 | 1/100 | 3/100 | 1/100 | 0/100 | |
| 2026-10-07 | 100/100 | 95/100 | 8/15 | 2/100 | 2/100 | 1/100 | 0/100 | changed |
| 2026-10-06 | 0/100 | — | — | — | — | — | — | |
| 2026-10-05 | 100/100 | 77/100 | 8/15 | 15/100 | 8/100 | 0/100 | 0/100 | |
| 2026-10-04 | 100/100 | 78/100 | 8/15 | 13/100 | 7/100 | 2/100 | 0/100 | |
| 2026-10-03 | 100/100 | 73/100 | 8/15 | 15/100 | 11/100 | 1/100 | 0/100 | |
| 2026-10-02 | 100/100 | 75/100 | 8/15 | 18/100 | 5/100 | 2/100 | 0/100 | |
| 2026-10-01 | 100/100 | 78/100 | 8/15 | 18/100 | 4/100 | 0/100 | 0/100 | |
| 2026-09-30 | 100/100 | 75/100 | 8/15 | 17/100 | 7/100 | 1/100 | 0/100 | |
| 2026-09-29 | 100/100 | 72/100 | 8/15 | 19/100 | 7/100 | 2/100 | 0/100 | |
| 2026-09-28 | 100/100 | 79/100 | 8/15 | 13/100 | 8/100 | 0/100 | 0/100 | |
| 2026-09-27 | 100/100 | 83/100 | 8/15 | 12/100 | 5/100 | 0/100 | 0/100 | |
| 2026-09-26 | 100/100 | 77/100 | 8/15 | 13/100 | 8/100 | 2/100 | 0/100 | |
| Date | Observed | Same answer twice | Refused (boundary) | Wobble: wording | answer | format | refusal | Change |
|---|
| 2026-10-10 | 100/100 | 49/100 | 7/15 | 20/100 | 27/100 | 3/100 | 1/100 | changed |
| 2026-10-09 | 100/100 | 47/100 | 7/15 | 19/100 | 31/100 | 2/100 | 1/100 | |
| 2026-10-08 | 100/100 | 47/100 | 7/15 | 20/100 | 27/100 | 4/100 | 2/100 | |
| Date | Observed | Same answer twice | Refused (boundary) | Wobble: wording | answer | format | refusal | Change |
|---|
| 2026-10-10 | 100/100 | 43/100 | 5/15 | 27/100 | 22/100 | 6/100 | 2/100 | |
| 2026-10-09 | 100/100 | 37/100 | 5/15 | 21/100 | 30/100 | 11/100 | 1/100 | |
| 2026-10-08 | 100/100 | 43/100 | 5/15 | 20/100 | 26/100 | 7/100 | 4/100 | |
| 2026-10-07 | 100/100 | 39/100 | 5/15 | 25/100 | 28/100 | 7/100 | 1/100 | |
| 2026-10-06 | 0/100 | — | — | — | — | — | — | |
| 2026-10-05 | 74/100 | 34/74 | — | 22/74 | 15/74 | 3/74 | 0/74 | |
| 2026-10-04 | 100/100 | 42/100 | 5/15 | 21/100 | 30/100 | 5/100 | 2/100 | |
| 2026-10-03 | 100/100 | 42/100 | 4/15 | 22/100 | 30/100 | 5/100 | 1/100 | |
| Date | Observed | Same answer twice | Refused (boundary) | Wobble: wording | answer | format | refusal | Change |
|---|
| 2026-10-03 | 100/100 | 41/100 | 6/15 | 16/100 | 34/100 | 9/100 | 0/100 | |
| 2026-10-02 | 100/100 | 40/100 | 7/15 | 17/100 | 29/100 | 13/100 | 1/100 | |
| 2026-10-01 | 100/100 | 44/100 | 5/15 | 13/100 | 29/100 | 12/100 | 2/100 | |
| 2026-09-30 | 100/100 | 38/100 | 6/15 | 18/100 | 33/100 | 10/100 | 1/100 | |
| 2026-09-29 | 100/100 | 39/100 | 5/15 | 14/100 | 34/100 | 11/100 | 2/100 | |
| 2026-09-28 | 100/100 | 39/100 | 8/15 | 18/100 | 33/100 | 7/100 | 3/100 | |
| 2026-09-27 | 100/100 | 37/100 | 7/15 | 19/100 | 32/100 | 11/100 | 1/100 | |
| 2026-09-26 | 100/100 | 43/100 | 6/15 | 19/100 | 23/100 | 13/100 | 2/100 | |
| Date | Observed | Same answer twice | Refused (boundary) | Wobble: wording | answer | format | refusal | Change |
|---|
| 2026-10-10 | 100/100 | 37/100 | 4/15 | 26/100 | 33/100 | 4/100 | 0/100 | |
| 2026-10-09 | 100/100 | 35/100 | 5/15 | 29/100 | 31/100 | 5/100 | 0/100 | |
| 2026-10-08 | 100/100 | 36/100 | 5/15 | 32/100 | 28/100 | 3/100 | 1/100 | |
| 2026-10-07 | 100/100 | 41/100 | 4/15 | 30/100 | 24/100 | 4/100 | 1/100 | |
| 2026-10-06 | 0/100 | — | — | — | — | — | — | |
| 2026-10-05 | 100/100 | 37/100 | 5/15 | 26/100 | 30/100 | 5/100 | 2/100 | |
| 2026-10-04 | 100/100 | 32/100 | 5/15 | 32/100 | 31/100 | 4/100 | 1/100 | |
| 2026-10-03 | 100/100 | 36/100 | 4/15 | 32/100 | 24/100 | 7/100 | 1/100 | |
The data is licensed under CC BY 4.0. Please credit https://aimodelrecord.com when you use it.