gpt4 · Model identity and integrity report

Name: gpt4 relay integrity probe aggregate
Creator: BazaarLink

Last updated：Jul 30, 2026, 12:49 PM · Data window: past 90 days

Measured results (90 days)

9 runs · 8 hosts · 7 anomalies

Family-fingerprint consistency

  openai           7  ████████████████████░░░░░   78% (suspicious)
  unknown          2  ██████░░░░░░░░░░░░░░░░░░░   22% (suspicious)

Observed V3 sub-model mapping

  → unknown                                  4 (44%) ⚠ mismatch
  → openai/gpt-5.4                           2 (22%) ⚠ mismatch sim=1.00
  → openai/gpt-4o                            1 (11%) ⚠ mismatch sim=0.82
  → openai/gpt-4.1                           1 (11%) ⚠ mismatch sim=0.74
  → openai/gpt-5.2                           1 (11%) ⚠ mismatch sim=0.90

Capability score distribution (auxiliary)

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

Host integrity ranking

Host	Runs	Average score	Family match	Most common V3 match
https://api.shubiaobiao.cn/v1	2	5	0%	—
https://api.vectorengine.cn/v1	1	78	0%	openai/gpt-4o (1)
https://runanytime.hxi.me/v1	1	0	0%	—
https://aiok.club/v1	1	86	0%	openai/gpt-4.1 (1)
https://api.qlhazycoder.top/v1	1	89	0%	openai/gpt-5.2 (1)
https://yunwu.ai/v1	1	0	0%	—
https://api.lingshuai.cc/v1	1	89	0%	openai/gpt-5.4 (1)
https://vip.auto-code.net/v1	1	89	0%	openai/gpt-5.4 (1)

Recent anomalies

https://api.vectorengine.cn/v1Jul 29, 2026, 12:05 PM

Self-claimed model: gpt-4
V2 predicted family: openai
V3 best match: openai/gpt-4o (sim 0.82)
Verdict: Full swap

https://runanytime.hxi.me/v1Jul 26, 2026, 3:47 PM

Self-claimed model: gpt-4
V2 predicted family: openai
V3 best match: —
Verdict: Family mismatch

https://aiok.club/v1Jun 19, 2026, 1:43 AM

Self-claimed model: GPT-4
V2 predicted family: openai
V3 best match: openai/gpt-4.1 (sim 0.74)
Verdict: Full swap

https://api.qlhazycoder.top/v1Jun 16, 2026, 1:37 AM

Self-claimed model: gpt-4
V2 predicted family: openai
V3 best match: openai/gpt-5.2 (sim 0.90)
Verdict: Full swap

https://yunwu.ai/v1Jun 15, 2026, 6:55 AM

Self-claimed model: gpt-4
V2 predicted family: openai
V3 best match: —
Verdict: Family mismatch

https://api.lingshuai.cc/v1May 6, 2026, 10:43 AM

Self-claimed model: gpt-4
V2 predicted family: openai
V3 best match: openai/gpt-5.4 (sim 1.00)
Verdict: Full swap

https://vip.auto-code.net/v1May 6, 2026, 3:14 AM

Self-claimed model: gpt-4
V2 predicted family: openai
V3 best match: openai/gpt-5.4 (sim 1.00)
Verdict: Full swap

Methodology

BazaarLink Probe combines a Bayesian V2 family classifier across nine fingerprint categories with V3 sub-model references to verify endpoints claiming gpt4. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →