← Back to model

agnes20flash · Model identity and integrity report

Last updated：Sep 13, 2026, 9:22 PM · Data window: past 90 days

Measured results (90 days)

4 runs · 1 hosts · 4 anomalies

Family-fingerprint consistency

  openai           4  █████████████████████████  100% (suspicious)

Observed V3 sub-model mapping

  → abstain                                  3 (75%) — low confidence
  → openai/gpt-5.5                           1 (25%) ⚠ mismatch sim=0.62

Capability score distribution (auxiliary)

  0-49     ██████░░░░░░░░░░░░░░░░░░░    1
  50-69    ██████░░░░░░░░░░░░░░░░░░░    1
  70-89    █████████████░░░░░░░░░░░░    2
  90-100   ░░░░░░░░░░░░░░░░░░░░░░░░░    0

Host integrity ranking

Host	Runs	Average score	Family match	Most common V3 match
https://apihub.agnes-ai.com/v1	4	52	0%	openai/gpt-5.5 (1)

Recent anomalies

https://apihub.agnes-ai.com/v1Jul 19, 2026, 2:34 PM

Self-claimed model: agnes-2.0-flash
V2 predicted family: openai
V3 best match: —
Verdict: V3 low confidence · Abstain

https://apihub.agnes-ai.com/v1Jul 13, 2026, 1:13 PM

Self-claimed model: agnes-2.0-flash
V2 predicted family: openai
V3 best match: —
Verdict: V3 low confidence · Abstain

https://apihub.agnes-ai.com/v1Jun 30, 2026, 2:12 AM

Self-claimed model: agnes-2.0-flash
V2 predicted family: openai
V3 best match: openai/gpt-5.5 (sim 0.62)
Verdict: Full swap

https://apihub.agnes-ai.com/v1Jun 26, 2026, 3:50 AM

Self-claimed model: agnes-2.0-flash
V2 predicted family: openai
V3 best match: —
Verdict: V3 low confidence · Abstain

Methodology

BazaarLink Probe combines a Bayesian V2 family classifier across nine fingerprint categories with V3 sub-model references to verify endpoints claiming agnes20flash. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →