Model checks
Vector Studios, as the auditors found it. Seller →
last audit 7 Nov · evidence checked on-chain×1.08
on its $ANTS points now
Approved auditors check that each seller serves the model it sells.
- 🧪They buy from it like any buyer: ~100 obscure number questions, compared with the real model.
- 📏The real model misses some too: that is the noise. A seller fails only if it misses clearly more, and two auditors must agree.
- 💰Each pass +2% on $ANTS points, up to +50%. Fails cost more: 1 −4%, 2 −16%, 3 −36%. Checked again every week.
How this works (Antseed's announcement ↗)
Approved auditors check that each seller serves the model it sells.
- 🧪They buy from it like any buyer: ~100 obscure number questions, compared with the real model.
- 📏The real model misses some too: that is the noise. A seller fails only if it misses clearly more, and two auditors must agree.
- 💰Each pass +2% on $ANTS points, up to +50%. Fails cost more: 1 −4%, 2 −16%, 3 −36%. Checked again every week.
Passed
4
SAME · counting now
Last audit
7 Nov
epoch 30 · counts to 31
| Model | Differed↑ | × noise | Fail line | Noise | Proof |
|---|
| GPT-6.1 SolSAME auditors split: 1 DIFF, 1 SAME | 1% | 2.0× | 11 under | 0.0% | ↗ |
| Kimi K3.1 ProSAME | 3% | 0.3× | 27 under | 8.2% | ↗ |
| GPT-6 AstraSAME | 5% | 1.0× | 19 under | 5.5% | ↗ |
| GLM 5.3SAME auditors split: 1 DIFF, 1 SAME | 11% | 0.9× | 25 under | 11.5% | ↗ |
| Kimi K3 1 DIFF report, a 2nd decides | 29% | 3.4× | +8 over | 8.7% | ↗ |
Kimi K31 DIFF
44 of 150 differed (fails from 36) · the real model differs from itself on 9%
Where it missed
Chemistry (melting points)5/7
Chinese literature (publication years)7/12
Chinese history (specific dates)4/10
Chinese internet (platform dates)1/3
For example
land area of toli county xinjiang in square kilometersreal 20,000 · seller 52,498
Diploid chromosome count of Botrychium virginianumreal 90 · seller 184
Diploid chromosome count of the fern *Psilotum nudum*real 104 · seller 208
✓Reasoning switched off with reasoning effort none, as the real model was asked.
| Round | Passed | Failed | Multiplier |
|---|
| 7 Novepoch 30 | 4 | 0+1 waiting | ×1.08 |
| 4 Novepoch 29 | 1 | 0 | ×1.02 |