🐜AntSeedStats
HomeNetworkModelsProvidersStakingFree AI$ANTS

Explore

OverviewRevenueProvidersBuyersModelsDistributionStakingRewardsDIEM$ANTSChannelsTransactionsLabs

Project

AboutFAQDevelopersFor agentsChangelogStatus

Antseed

WebsiteDownload Antseed DesktopDocsGitHubX (@antseedai)Telegram
🐜AntSeedStats
The metrics and intelligence layer of the Antseed ecosystem.
Built by @gekko_eth
AntSeedStats is an independent, community-built dashboard, not officially affiliated with the Antseed team. All data is sourced on-chain and from public endpoints.

Model checks

Apiport, as the auditors found it. Provider →

last audit 18 Nov · evidence checked on-chain
×0.96

on its $ANTS points now

How this works (Antseed's announcement ↗)

Approved auditors check that each provider 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 provider 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 provider 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 provider 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
0
SAME · counting now
Failed
1
DIFF
Last audit
18 Nov
epoch 31 · counts to 32
By model
ModelDiffered↑× noiseFail lineNoiseProof
GPT-6 AstraDIFF51%51×+37 over1.0%↗
Why it failed
GPT-6 AstraDIFF

31 of 110 differed (fails from 21) · 51 of 100 differed (fails from 14) · the real model differs from itself on 1%

Where it missed
Chinese literature (publication years)9/11
Biology (chromosome numbers)14/21
Medical (drug half-lives)15/24
Chinese history (specific dates)3/5
For example
Nedosiran terminal plasma half-life in hoursreal 15 · provider 153
Sichuan Garze County total area (square kilometres)real 7,357 · provider 14,785
NGC 1514 central-binary orbital period reported by Jones et al. (2017) (days)real 3,306 · provider 3.161

✓Reasoning switched off with reasoning effort none, as the real model was asked.

audited 18 Nov, 18:14 UTC · epoch 31
test power 97% · median answer 12.9 s · 2 audits
4 older checks · 1 failed · not counting now
ModelDiffered↑× noiseFail lineNoiseProof
Claude Sonnet 5.5SAME 21 Oct · first round, before points11%1.1×12 under9.8%↗
Kimi K3SAME 21 Oct · first round, before points16%1.7×4 under9.0%↗
GPT-6.1 SolDIFF 7 Nov · expired19%38×+7 over0.0%↗
Claude Opus 5.5 21 Oct · not tested · provider unreachable————↗
GPT-6.1 SolDIFFno longer counts

19 of 100 differed (fails from 12) · 18 of 99 differed (fails from 12) · the real model differs from itself on 0%

Where it missed
Chemistry (melting points)1/3
Chemistry (boiling points)1/3
Programming (release years)5/16
Pop culture (box office and ratings)7/24
For example
Phenoxathiin melting point (°C)real 56 · provider 155
Salvinia cucullata somatic chromosome count (2n, chromosomes)real 18 · provider 45
Anqing Yangtze River Highway Bridge main span length (metres)real 510 · provider 1,200

✓Reasoning switched off with reasoning effort none, as the real model was asked.

audited 7 Nov, 22:28 UTC · epoch 30
test power 98% · median answer 23.9 s · 2 audits
Audit rounds
RoundPassedFailedNot testedMultiplier
18 Novepoch 31010×0.84
7 Novepoch 30020×0.84
4 Novepoch 29020×0.84
21 Octepoch 27 · before points401—

Audited by approved auditors · evidence on IPFS, hash-checked on-chain.