openai/gpt-4o · Model identity and integrity report

Name: openai/gpt-4o relay integrity probe aggregate
Creator: BazaarLink

Last updated：Jul 30, 2026, 12:08 PM · Data window: past 90 days

Verification capability — fingerprint reference available

Family: openai
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: 52 weighted rules · 8 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)

17 runs · 10 hosts · 3 anomalies

Family-fingerprint consistency

  openai          14  █████████████████████░░░░   82% (expected)
  anthropic        1  ██░░░░░░░░░░░░░░░░░░░░░░░    6% (suspicious)
  unknown          1  ██░░░░░░░░░░░░░░░░░░░░░░░    6% (suspicious)
  deepseek         1  ██░░░░░░░░░░░░░░░░░░░░░░░    6% (suspicious)

Observed V3 sub-model mapping

  → openai/gpt-4o                           11 (65%) ✓ correct sim=0.89
  → abstain                                  2 (12%) — low confidence
  → unknown                                  2 (12%) ⚠ mismatch
  → deepseek/deepseek-v4-flash               1 ( 6%) ⚠ mismatch sim=0.98
  → openai/gpt-4.1                           1 ( 6%) ⚠ mismatch sim=0.79

Capability score distribution (auxiliary)

  0-49     ████████████░░░░░░░░░░░░░    8
  50-69    ░░░░░░░░░░░░░░░░░░░░░░░░░    0
  70-89    ██████░░░░░░░░░░░░░░░░░░░    4
  90-100   ███████░░░░░░░░░░░░░░░░░░    5

Host integrity ranking

Host	Runs	Average score	Family match	Most common V3 match
https://www.moyu.info/v1	6	85	100%	openai/gpt-4o (5)
https://yinli.pro/v1	3	62	100%	openai/gpt-4o (3)
https://alloy-liver-congressional-role.trycloudflare.com/v1	1	0	0%	—
https://httpbin.org/redirect-to?url=http://127.0.0.1:80	1	3	0%	—
https://runanytime.hxi.me/v1	1	0	100%	—
https://sub2api.crazybunqnq.com/v1	1	0	0%	deepseek/deepseek-v4-flash (1)
https://api.openai-hub.net/v1	1	76	100%	openai/gpt-4o (1)
https://api.kourichat.com/v1	1	0	100%	openai/gpt-4o (1)
https://yunwu.ai/v1	1	0	100%	—
https://api.shubiaobiao.cn/v1	1	0	100%	openai/gpt-4o (1)

Recent anomalies

https://alloy-liver-congressional-role.trycloudflare.com/v1Jul 28, 2026, 5:03 AM

Self-claimed model: gpt-4o
V2 predicted family: anthropic
V3 best match: —
Verdict: V3 low confidence · Abstain

https://sub2api.crazybunqnq.com/v1Jul 25, 2026, 5:15 PM

Self-claimed model: gpt-4o
V2 predicted family: deepseek
V3 best match: deepseek/deepseek-v4-flash (sim 0.98)
Verdict: Full swap

https://www.moyu.info/v1May 20, 2026, 3:44 AM

Self-claimed model: GPT-4o
V2 predicted family: openai
V3 best match: openai/gpt-4.1 (sim 0.79)
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 openai/gpt-4o. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →