openai/gpt-image-2 · Model identity and integrity report
Last updated：Jul 30, 2026, 1:41 PM · Data window: past 90 days
Measured results (90 days)
10 runs · 9 hosts · 6 anomalies
Family-fingerprint consistency
openai 5 █████████████░░░░░░░░░░░░ 50% (expected) google 2 █████░░░░░░░░░░░░░░░░░░░░ 20% (suspicious) unknown 2 █████░░░░░░░░░░░░░░░░░░░░ 20% (suspicious) anthropic 1 ███░░░░░░░░░░░░░░░░░░░░░░ 10% (suspicious)
Observed V3 sub-model mapping
→ unknown 6 (60%) ⚠ mismatch → abstain 4 (40%) — low confidence
Capability score distribution (auxiliary)
0-49 █████████████████████████ 10 50-69 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0 70-89 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0 90-100 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://api.rkai6.com/v1 | 2 | 23 | 50% | — |
| https://nexaxis.ai/v1 | 1 | 0 | 100% | — |
| https://api.1sfpk.com/v1 | 1 | 0 | 100% | — |
| http://114.132.240.246/v1 | 1 | 20 | 0% | — |
| http://154.17.19.146:3000/v1 | 1 | 6 | 100% | — |
| https://yunwu.ai/v1 | 1 | 0 | 100% | — |
| https://api.sccens.net/v1 | 1 | 0 | 0% | — |
| https://api.manxiaobai.online/v1 | 1 | 0 | 0% | — |
| https://www.moyu.info/v1 | 1 | 6 | 0% | — |
Recent anomalies
http://114.132.240.246/v1Jun 18, 2026, 6:59 AM
- Self-claimed model
- gpt-image-2
- V2 predicted family
- V3 best match
- —
- Verdict
- Family mismatch
http://154.17.19.146:3000/v1Jun 11, 2026, 4:42 PM
- Self-claimed model
- gpt-image-2
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.rkai6.com/v1Jun 10, 2026, 7:18 AM
- Self-claimed model
- gpt-image-2
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.rkai6.com/v1Jun 9, 2026, 8:53 AM
- Self-claimed model
- gpt-image-2
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://yunwu.ai/v1Jun 5, 2026, 10:43 AM
- Self-claimed model
- gpt-image-2
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.manxiaobai.online/v1May 24, 2026, 10:44 AM
- Self-claimed model
- gpt-image-2
- 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 openai/gpt-image-2. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.