← Back to model

grok45high · Model identity and integrity report

Name: grok45high relay integrity probe aggregate
Creator: BazaarLink

Last updated：Aug 19, 2026, 5:18 AM · Data window: past 90 days

Measured results (90 days)

6 runs · 3 hosts · 5 anomalies

Family-fingerprint consistency

  xai              4  █████████████████░░░░░░░░   67% (suspicious)
  unknown          1  ████░░░░░░░░░░░░░░░░░░░░░   17% (suspicious)
  openai           1  ████░░░░░░░░░░░░░░░░░░░░░   17% (suspicious)

Observed V3 sub-model mapping

  → x-ai/grok-4.3                            2 (33%) ⚠ mismatch sim=0.72
  → abstain                                  2 (33%) — low confidence
  → unknown                                  1 (17%) ⚠ mismatch
  → x-ai/grok-4.5                            1 (17%) ⚠ mismatch sim=0.80

Capability score distribution (auxiliary)

  0-49     ████░░░░░░░░░░░░░░░░░░░░░    1
  50-69    ████████░░░░░░░░░░░░░░░░░    2
  70-89    █████████████░░░░░░░░░░░░    3
  90-100   ░░░░░░░░░░░░░░░░░░░░░░░░░    0

Host integrity ranking

Host	Runs	Average score	Family match	Most common V3 match
https://routifyapi.com/v1	3	71	0%	x-ai/grok-4.3 (1)
https://api.ngawi-proxy.cloud/v1	2	33	0%	x-ai/grok-4.3 (1)
https://mainproject.my.id/v1	1	71	0%	—

Recent anomalies

https://api.ngawi-proxy.cloud/v1Aug 19, 2026, 4:22 AM

Self-claimed model: grok-4.5-high
V2 predicted family: xai
V3 best match: x-ai/grok-4.3 (sim 0.71)
Verdict: Full swap

https://routifyapi.com/v1Aug 13, 2026, 10:56 AM

Self-claimed model: grok-4.5-high
V2 predicted family: xai
V3 best match: x-ai/grok-4.3 (sim 0.74)
Verdict: Full swap

https://routifyapi.com/v1Aug 10, 2026, 4:31 PM

Self-claimed model: grok-4.5-high
V2 predicted family: xai
V3 best match: —
Verdict: V3 low confidence · Abstain

https://routifyapi.com/v1Aug 10, 2026, 3:40 PM

Self-claimed model: grok-4.5-high
V2 predicted family: xai
V3 best match: x-ai/grok-4.5 (sim 0.80)
Verdict: Full swap

https://mainproject.my.id/v1Jul 16, 2026, 3:23 AM

Self-claimed model: gcli/grok-4.5-high
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 grok45high. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →