gemini25pro · Model identity and integrity report
Last updated：Sep 13, 2026, 9:22 PM · Data window: past 90 days
Measured results (90 days)
30 runs · 18 hosts · 27 anomalies
Family-fingerprint consistency
google 22 ██████████████████░░░░░░░ 73% (suspicious) anthropic 5 ████░░░░░░░░░░░░░░░░░░░░░ 17% (suspicious) unknown 3 ███░░░░░░░░░░░░░░░░░░░░░░ 10% (suspicious)
Observed V3 sub-model mapping
→ google/gemini-2.5-pro 11 (37%) ⚠ mismatch sim=0.92 → abstain 10 (33%) — low confidence → unknown 5 (17%) ⚠ mismatch → google/gemini-3.1-pro-preview 2 ( 7%) ⚠ mismatch sim=0.80 → anthropic/claude-sonnet-4.5 1 ( 3%) ⚠ mismatch sim=0.93 → google/gemini-3.1-flash-lite-preview 1 ( 3%) ⚠ mismatch sim=0.92
Capability score distribution (auxiliary)
0-49 █████████████████████░░░░ 25 50-69 ███░░░░░░░░░░░░░░░░░░░░░░ 4 70-89 █░░░░░░░░░░░░░░░░░░░░░░░░ 1 90-100 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://nexaxis.ai/v1 | 7 | 0 | 0% | google/gemini-2.5-pro (4) |
| https://laisiai.tw.cpolar.io/v1 | 3 | 22 | 0% | — |
| https://speed.toter.me/v1 | 3 | 19 | 0% | anthropic/claude-sonnet-4.5 (1) |
| https://tian-shu.org/v1 | 2 | 0 | 0% | google/gemini-2.5-pro (1) |
| https://seesaw2ai.com/v1 | 2 | 0 | 0% | google/gemini-2.5-pro (1) |
| https://speed9.toter.me/v1 | 1 | 0 | 0% | — |
| https://ckffai.com/v1 | 1 | 0 | 0% | — |
| https://api.easytransnote.com/beta/v1 | 1 | 77 | 0% | — |
| https://nodexapi.it.com/v1 | 1 | 0 | 0% | — |
| https://www.anyfast.ai/v1 | 1 | 57 | 0% | google/gemini-2.5-pro (1) |
| https://api.relaycore.cn/v1 | 1 | 0 | 0% | — |
| https://api.yuegle.com/v1 | 1 | 0 | 0% | google/gemini-2.5-pro (1) |
| https://gemini.beijixingxing.com/v1 | 1 | 50 | 0% | google/gemini-2.5-pro (1) |
| https://api.ekan8.com/v1 | 1 | 0 | 0% | google/gemini-2.5-pro (1) |
| https://quotarouter.ai/v1 | 1 | 0 | 0% | — |
| https://yunwu.ai/v1 | 1 | 0 | 0% | google/gemini-3.1-pro-preview (1) |
| https://api.toknex.ai/v1 | 1 | 0 | 0% | google/gemini-2.5-pro (1) |
| https://goodreamapi.online/v1 | 1 | 16 | 0% | — |
Recent anomalies
https://speed9.toter.me/v1Sep 12, 2026, 8:30 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- Full swap
https://laisiai.tw.cpolar.io/v1Sep 6, 2026, 1:30 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://laisiai.tw.cpolar.io/v1Sep 5, 2026, 9:29 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://laisiai.tw.cpolar.io/v1Sep 5, 2026, 9:27 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- Family mismatch
https://speed.toter.me/v1Sep 2, 2026, 7:12 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://speed.toter.me/v1Aug 28, 2026, 3:14 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- Full swap
https://api.easytransnote.com/beta/v1Aug 15, 2026, 5:37 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://nodexapi.it.com/v1Aug 15, 2026, 5:13 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- Family mismatch
https://speed.toter.me/v1Aug 5, 2026, 11:38 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-4.5 (sim 0.93)
- Verdict
- Full swap
https://www.anyfast.ai/v1Aug 4, 2026, 3:25 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-2.5-pro (sim 0.90)
- Verdict
- Full swap
https://api.relaycore.cn/v1Aug 2, 2026, 11:43 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://tian-shu.org/v1Aug 2, 2026, 5:57 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-2.5-pro (sim 0.82)
- Verdict
- Full swap
https://tian-shu.org/v1Aug 2, 2026, 5:53 PM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- Family mismatch
https://nexaxis.ai/v1Aug 2, 2026, 6:36 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-3.1-flash-lite-preview (sim 0.92)
- Verdict
- Full swap
https://nexaxis.ai/v1Aug 2, 2026, 6:35 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-3.1-pro-preview (sim 0.80)
- Verdict
- Full swap
https://api.yuegle.com/v1Jul 26, 2026, 10:49 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-2.5-pro (sim 0.89)
- Verdict
- Full swap
https://gemini.beijixingxing.com/v1Jul 9, 2026, 9:00 PM
- Self-claimed model
- 假流式-gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-2.5-pro (sim 0.91)
- Verdict
- Full swap
https://api.ekan8.com/v1Jul 3, 2026, 6:22 PM
- Self-claimed model
- 云雾-gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-2.5-pro (sim 0.85)
- Verdict
- Full swap
https://nexaxis.ai/v1Jun 30, 2026, 10:43 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- google/gemini-2.5-pro (sim 1.00)
- Verdict
- Full swap
https://quotarouter.ai/v1Jun 30, 2026, 8:22 AM
- Self-claimed model
- gemini-2.5-pro
- V2 predicted family
- V3 best match
- —
- Verdict
- Family mismatch
Methodology
BazaarLink Probe combines a Bayesian V2 family classifier across nine fingerprint categories with V3 sub-model references to verify endpoints claiming gemini25pro. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.