← Back to model

gemini35flash · Model identity and integrity report

Last updated：Sep 13, 2026, 9:15 PM · Data window: past 90 days

Measured results (90 days)

19 runs · 14 hosts · 15 anomalies

Family-fingerprint consistency

  google          12  ████████████████░░░░░░░░░   63% (expected)
  unknown          4  █████░░░░░░░░░░░░░░░░░░░░   21% (suspicious)
  anthropic        2  ███░░░░░░░░░░░░░░░░░░░░░░   11% (suspicious)
  openai           1  █░░░░░░░░░░░░░░░░░░░░░░░░    5% (suspicious)

Observed V3 sub-model mapping

  → abstain                                  8 (42%) — low confidence
  → google/gemini-3.1-pro-preview            5 (26%) ⚠ mismatch sim=0.83
  → unknown                                  4 (21%) ⚠ mismatch
  → google/gemini-3.6-flash                  1 ( 5%) ⚠ mismatch sim=0.70
  → anthropic/claude-sonnet-4.6              1 ( 5%) ⚠ mismatch sim=0.70

Capability score distribution (auxiliary)

  0-49     ████████████░░░░░░░░░░░░░    9
  50-69    ████████░░░░░░░░░░░░░░░░░    6
  70-89    █████░░░░░░░░░░░░░░░░░░░░    4
  90-100   ░░░░░░░░░░░░░░░░░░░░░░░░░    0

Host integrity ranking

Host	Runs	Average score	Family match	Most common V3 match
https://api.tokeneasy.com.cn/v1	3	0	33%	anthropic/claude-sonnet-4.6 (1)
https://openrouter.ai/api/v1	3	64	100%	google/gemini-3.1-pro-preview (1)
https://www.anyfast.ai/v1	2	62	100%	—
https://laisiai.tw.cpolar.io/v1	1	0	0%	—
https://image.apiroute.top/v1	1	43	100%	—
https://bboluo.com/v1	1	0	0%	—
https://api.aicodemirror.ai/api/gemini/v1	1	79	100%	google/gemini-3.1-pro-preview (1)
https://api.blackbox.ai/v1	1	66	100%	—
https://lanxiu.cc/v1	1	76	0%	google/gemini-3.6-flash (1)
https://llm.teitoku2.cc.cd/v1	1	0	0%	—
https://api.xgapi.top/v1	1	80	100%	google/gemini-3.1-pro-preview (1)
https://api.ekan8.com/v1	1	0	0%	—
https://api.api2cn.com/v1	1	0	100%	google/gemini-3.1-pro-preview (1)
https://yunwu.ai/v1	1	80	100%	google/gemini-3.1-pro-preview (1)

Recent anomalies

https://image.apiroute.top/v1Sep 2, 2026, 1:50 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://bboluo.com/v1Aug 11, 2026, 12:58 AM

Self-claimed model: (反重力)gemini-3.5-flash
V2 predicted family: openai
V3 best match: —
Verdict: Full swap

https://api.aicodemirror.ai/api/gemini/v1Aug 8, 2026, 9:21 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: google/gemini-3.1-pro-preview (sim 0.96)
Verdict: Sub-model downgrade

https://api.blackbox.ai/v1Aug 6, 2026, 2:48 PM

Self-claimed model: blackboxai/google/gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://www.anyfast.ai/v1Aug 4, 2026, 3:52 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://www.anyfast.ai/v1Aug 4, 2026, 2:51 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://lanxiu.cc/v1Aug 4, 2026, 1:54 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: anthropic
V3 best match: google/gemini-3.6-flash (sim 0.70)
Verdict: Sub-model downgrade

https://api.tokeneasy.com.cn/v1Jul 28, 2026, 6:14 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://api.tokeneasy.com.cn/v1Jul 28, 2026, 6:13 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: anthropic
V3 best match: anthropic/claude-sonnet-4.6 (sim 0.70)
Verdict: Full swap

https://openrouter.ai/api/v1Jul 22, 2026, 11:00 AM

Self-claimed model: google/gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://openrouter.ai/api/v1Jul 22, 2026, 10:59 AM

Self-claimed model: google/gemini-3.5-flash
V2 predicted family: google
V3 best match: google/gemini-3.1-pro-preview (sim 0.94)
Verdict: Sub-model downgrade

https://openrouter.ai/api/v1Jul 22, 2026, 10:58 AM

Self-claimed model: google/gemini-3.5-flash
V2 predicted family: google
V3 best match: —
Verdict: V3 low confidence · Abstain

https://api.xgapi.top/v1Jun 30, 2026, 12:04 PM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: google/gemini-3.1-pro-preview (sim 0.78)
Verdict: Sub-model downgrade

https://api.api2cn.com/v1Jun 20, 2026, 4:16 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: google/gemini-3.1-pro-preview (sim 0.72)
Verdict: Sub-model downgrade

https://yunwu.ai/v1Jun 16, 2026, 5:38 AM

Self-claimed model: gemini-3.5-flash
V2 predicted family: google
V3 best match: google/gemini-3.1-pro-preview (sim 0.77)
Verdict: Sub-model downgrade

Methodology

BazaarLink Probe combines a Bayesian V2 family classifier across nine fingerprint categories with V3 sub-model references to verify endpoints claiming gemini35flash. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →