claudesonnet5thinking · Model identity and integrity report
Last updated：Sep 11, 2026, 11:53 AM · Data window: past 90 days
Measured results (90 days)
5 runs · 5 hosts · 5 anomalies
Family-fingerprint consistency
anthropic 4 ████████████████████░░░░░ 80% (suspicious) openai 1 █████░░░░░░░░░░░░░░░░░░░░ 20% (suspicious)
Observed V3 sub-model mapping
→ abstain 4 (80%) — low confidence → anthropic/claude-opus-4.7 1 (20%) ⚠ mismatch sim=0.71
Capability score distribution (auxiliary)
0-49 ███████████████░░░░░░░░░░ 3 50-69 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0 70-89 ██████████░░░░░░░░░░░░░░░ 2 90-100 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://api.sqface.sbs/v1 | 1 | 0 | 0% | — |
| https://api.9aws.net/v1 | 1 | 0 | 0% | — |
| https://htai91.com/v1 | 1 | 0 | 0% | — |
| https://apibasis.com/v1 | 1 | 84 | 0% | — |
| https://api.oaibest.com/v1 | 1 | 84 | 0% | anthropic/claude-opus-4.7 (1) |
Recent anomalies
https://api.sqface.sbs/v1Sep 8, 2026, 10:19 PM
- Self-claimed model
- claude-sonnet-5-thinking
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.9aws.net/v1Aug 23, 2026, 12:51 PM
- Self-claimed model
- claude-sonnet-5-thinking
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://htai91.com/v1Aug 22, 2026, 4:45 PM
- Self-claimed model
- m365/claude-sonnet-5-thinking
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://apibasis.com/v1Aug 3, 2026, 9:47 AM
- Self-claimed model
- claude-sonnet-5-thinking
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.oaibest.com/v1Jul 13, 2026, 5:45 AM
- Self-claimed model
- claude-sonnet-5-thinking
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-opus-4.7 (sim 0.71)
- 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 claudesonnet5thinking. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.