gemini31flashlitepreview · Model identity and integrity report
Last updated：Sep 13, 2026, 5:28 PM · Data window: past 90 days
Verification capability — fingerprint reference available
- Family
- V3 sub-model reference
- Available ✓
- Checks
- 知識截止 · 能力探針 · 拒絕模板 · 拒絕溫度參數
- Family-fingerprint coverage
- 17 weighted rules · 5 feature categories
Concrete reference answers, refusal templates, and individual signal weights are not published because disclosure would help spoofing endpoints evade detection.
Measured results (90 days)
9 runs · 7 hosts · 8 anomalies
Family-fingerprint consistency
openai 5 ██████████████░░░░░░░░░░░ 56% (suspicious) anthropic 3 ████████░░░░░░░░░░░░░░░░░ 33% (suspicious) unknown 1 ███░░░░░░░░░░░░░░░░░░░░░░ 11% (suspicious)
Observed V3 sub-model mapping
→ abstain 8 (89%) — low confidence → unknown 1 (11%) ⚠ mismatch
Capability score distribution (auxiliary)
0-49 ██████████████░░░░░░░░░░░ 5 50-69 ███░░░░░░░░░░░░░░░░░░░░░░ 1 70-89 ████████░░░░░░░░░░░░░░░░░ 3 90-100 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://www.anyfast.ai/v1 | 2 | 77 | 0% | — |
| https://seesaw2ai.com/v1 | 2 | 0 | 0% | — |
| https://api.gemai.cc/v1 | 1 | 0 | 0% | — |
| https://yjapi.manqiaotechnology.com/v1 | 1 | 0 | 0% | — |
| https://image.apiroute.top/v1 | 1 | 70 | 0% | — |
| https://rumahai.net/api/v1 | 1 | 61 | 0% | — |
| https://api.yuegle.com/v1 | 1 | 0 | 0% | — |
Recent anomalies
https://api.gemai.cc/v1Sep 4, 2026, 10:51 AM
- Self-claimed model
- [满血A]gemini-3.1-flash-lite-preview
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://yjapi.manqiaotechnology.com/v1Sep 3, 2026, 5:41 AM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://image.apiroute.top/v1Sep 2, 2026, 1:57 AM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://rumahai.net/api/v1Aug 28, 2026, 6:54 PM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://www.anyfast.ai/v1Aug 4, 2026, 7:48 AM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- Full swap
https://www.anyfast.ai/v1Aug 4, 2026, 3:07 AM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- Full swap
https://api.yuegle.com/v1Jul 26, 2026, 10:51 AM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- Full swap
https://seesaw2ai.com/v1Jun 28, 2026, 8:43 AM
- Self-claimed model
- gemini-3.1-flash-lite-preview
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
Methodology
BazaarLink Probe combines a Bayesian V2 family classifier across nine fingerprint categories with V3 sub-model references to verify endpoints claiming gemini31flashlitepreview. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.