🐜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

antseed-cosmic-gecko-c46d, as the auditors found it. Provider →

last audit 7 Nov · evidence checked on-chain
×1

no check counts 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
0
DIFF
Last audit
7 Nov
epoch 30 · no longer counts
By model

No check counts now. Last audited 7 Nov.

8 older checks · 5 failed · not counting now
ModelDiffered↑× noiseFail lineNoiseProof
GPT-6 AstraSAME 7 Nov · expired · auditors split: 1 DIFF, 1 SAME2%1.0×15 under1.8%↗
Kimi K3.1 ProSAME 7 Nov · expired · auditors split: 1 DIFF, 1 SAME5%1.7×14 under2.7%↗
DeepSeek V4 ProSAME 7 Nov · expired · auditors split: 1 DIFF, 1 SAME11%1.4×18 under7.9%↗
GPT-6.1 SolDIFF 7 Nov · expired20%40×+8 over0.0%↗
Claude Sonnet 5.5DIFF 7 Nov · expired25%7.5×+9 over3.3%↗
Claude Opus 5.5DIFF 7 Nov · expired25%50×+13 over0.0%↗
Kimi K3DIFF 7 Nov · expired28%3.2×+6 over8.7%↗
GLM 5.3DIFF 7 Nov · expired58%5.0×+36 over11.5%↗
Kimi K3DIFFno longer counts

42 of 150 differed (fails from 36) · 36 of 150 differed (fails from 35) · the real model differs from itself on 9%

Where it missed
Chemistry (melting points)7/13
Medical (drug half-lives)10/22
Biology (chromosome numbers)8/24
Programming (release years)7/21
For example
Linnik constant upper bound (classical)real 5 · provider 550
area of wenquan county xinjiang in square kilometersreal 5,900 · provider 31,840
land area of toli county xinjiang in square kilometersreal 20,000 · provider 52,490

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

audited 7 Nov, 22:30 UTC · epoch 30
test power 96% · median answer 7.6 s · 2 audits
GPT-6.1 SolDIFFno longer counts

20 of 100 differed (fails from 12) · 16 of 99 differed (fails from 12) · the real model differs from itself on 0%

Where it missed
Programming (release years)6/16
Chinese internet (platform dates)7/21
Chinese history (specific dates)2/6
Chemistry (melting points)1/3
For example
NGC 4650A catalogued heliocentric redshift (dimensionless)real 0.009587 · provider 0.0397
Leptopteris laxa diploid chromosome number (2n, chromosomes)real 44 · provider 156
Leptopteris hymenophylloides diploid sporophyte chromosome number (2n, chromosomes)real 44 · provider 156

✓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 4.7 s · 2 audits
GLM 5.3DIFFno longer counts

50 of 130 differed (fails from 33) · 75 of 130 differed (fails from 39) · the real model differs from itself on 12%

Where it missed
Chemistry (melting points)12/16
Astronomy (periods and distances)10/16
Pop culture (box office and ratings)10/17
Chinese history (specific dates)7/12
For example
fifth bernoulli number b5 numeratorreal 0 · provider -1
year the Han dynasty sent Zhang Qian on his second western missionreal 119 · provider -119
melting point of Mg2Ni intermetallic in Kreal 760 · provider 1,723

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

audited 7 Nov, 22:26 UTC · epoch 30
test power 92% · median answer 5.6 s · 2 audits
Claude Sonnet 5.5DIFFno longer counts

25 of 100 differed (fails from 17) · 30 of 120 differed (fails from 21) · the real model differs from itself on 3%

Where it missed
Mathematics (special values)8/14
Chemistry (boiling points)7/16
Medical (drug half-lives)9/26
Biology (chromosome numbers)8/31
For example
Diploid chromosome number (2n) of Azolla filiculoides, countreal 44 · provider 132
Psilotum nudum diploid chromosome number (2n)real 104 · provider 210
Diploid chromosome number (2n) of Psilotum nudum, countreal 104 · provider 208

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

audited 7 Nov, 22:24 UTC · epoch 30
test power 96% · median answer 7.3 s · 2 audits
Claude Opus 5.5DIFFno longer counts

31 of 110 differed (fails from 19) · 25 of 100 differed (fails from 12) · the real model differs from itself on 0%

Where it missed
Programming (release years)9/17
Medical (drug half-lives)9/20
Pop culture (box office and ratings)5/12
Chinese literature (publication years)8/24
For example
shijing xiaoya section poems with extant text countreal 74 · provider 305
hymenophyllum tunbrigense diploid chromosome number 2n countreal 26 · provider 72
notothenia coriiceps diploid chromosome count 2nreal 22 · provider 48

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

audited 7 Nov, 22:08 UTC · epoch 30
test power 97% · median answer 19.2 s · 2 audits
Audit rounds
RoundPassedFailedNot testedMultiplier
7 Novepoch 30350×0.06
4 Novepoch 29061×0.70

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