Model checks
MiniMax M3, provider by provider. Model →
last audit 18 Nov · epoch 31 · evidence checked on-chain12 of 14
providers answered as the model
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
12
SAME
Failed
2
DIFF
Not tested
4
the audit could not run
By provider
Compared with minimax/minimax-m3 on OpenRouter · pinned to minimax · reasoning none
| Provider | Differed↑ | × noise | Fail line | Noise | Proof |
|---|---|---|---|---|---|
| ZLKPro-ApiSAME | 12% | 0.9× | 31 under | 14.1% | ↗ |
| Dark SignalSAME | 13% | 0.9× | 30 under | 14.1% | ↗ |
| Argus AISAME | 13% | 0.9× | 30 under | 14.1% | ↗ |
| attacompute.aiSAME | 14% | 1.0× | 28 under | 14.1% | ↗ |
| BestSAME | 14% | 1.0× | 28 under | 14.1% | ↗ |
| Vikey Seller no verdict this round | 14% | 1.0× | 8 under | 14.1% | ↗ |
| BabyCaiSAME | 15% | 1.0× | 27 under | 14.1% | ↗ |
| Edith AISAME | 16% | 1.1× | 25 under | 14.1% | ↗ |
| LLMartSAME | 17% | 1.2× | 23 under | 14.1% | ↗ |
| Unified Rack · DeepSeek 240 tok/sSAME | 18% | 1.3× | 22 under | 14.1% | ↗ |
| bartly.eth64.deSAME | 18% | 1.3× | 22 under | 14.1% | ↗ |
| Hana Gateway ✅SAME | 19% | 1.3× | 19 under | 14.1% | ↗ |
| dealerSAME | 24% | 1.7× | 12 under | 14.1% | ↗ |
| Open ForgeDIFF | 44% | 3.1× | +25 over | 14.4% | ↗ |
| Relay WorksDIFF | 48% | 3.3× | +31 over | 14.4% | ↗ |
| incharge-austin not tested · provider unreachable | — | — | — | — | ↗ |
| antseed-aggregator not tested · provider unreachable | — | — | — | — | ↗ |
| Venice.ai Proxy not tested · provider unreachable | — | — | — | — | ↗ |
5 older checks · 4 failed · not counting now
| Provider | Differed↑ | × noise | Fail line | Noise | Proof |
|---|---|---|---|---|---|
| Super Seeder 17 Nov · 1 DIFF report, no verdict | 47% | 3.4× | +28 over | 14.1% | ↗ |
| Fire Ant 🔥🐜DIFF 7 Nov · expired | 48% | 3.0× | +29 over | 16.1% | ↗ |
| ▲ Apex AntDIFF 7 Nov · expired | 49% | 3.1× | +31 over | 16.1% | ↗ |
| 💬 BanterDIFF 7 Nov · expired | 50% | 3.1× | +29 over | 16.1% | ↗ |
| klinnDIFF 7 Nov · expired | 55% | 3.4× | +37 over | 16.1% | ↗ |
Where providers miss
| Subject | Passed providers missed | Failed providers missed |
|---|---|---|
| Chinese literature (publication years) | 13% | 44% |
| Programming (release years) | 12% | 55% |
| Chinese geography (measurements) | 6% | 34% |
| Internet culture (dates and numbers) | 7% | 18% |
| Chinese history (specific dates) | 11% | 45% |
| Chinese internet (platform dates) | 10% | 35% |
| Mathematics (special values) | 36% | 33% |
| Physics (material properties) | 43% | 66% |
| Medical (drug half-lives) | 24% | 75% |
| Chemistry (melting points) | 9% | 47% |
Audit rounds
| Round | Passed | Failed | Not tested |
|---|---|---|---|
| 17 Novepoch 31 | 12 | 2+1 waiting | 4 |
| 7 Novepoch 30 | 12 | 4 | 4 |
Audited by approved auditors · evidence on IPFS, hash-checked on-chain.