← Back to model

qwen3824ta95b · Model identity and integrity report

Last updated：Aug 23, 2026, 5:53 AM · Data window: past 90 days

Measured results (90 days)

8 runs · 2 hosts · 6 anomalies

Family-fingerprint consistency

  qwen             7  ██████████████████████░░░   88% (expected)
  unknown          1  ███░░░░░░░░░░░░░░░░░░░░░░   13% (suspicious)

Observed V3 sub-model mapping

  → qwen/qwen3-max                           3 (38%) ⚠ mismatch sim=0.64
  → abstain                                  3 (38%) — low confidence
  → unknown                                  2 (25%) ⚠ mismatch

Capability score distribution (auxiliary)

  0-49     ████████████████░░░░░░░░░    5
  50-69    ██████████░░░░░░░░░░░░░░░    3
  70-89    ░░░░░░░░░░░░░░░░░░░░░░░░░    0
  90-100   ░░░░░░░░░░░░░░░░░░░░░░░░░    0

Host integrity ranking

Host	Runs	Average score	Family match	Most common V3 match
https://api.gnrt.dev/v1	6	27	83%	qwen/qwen3-max (1)
https://api-inference.bitdeer.ai/v1	2	0	100%	qwen/qwen3-max (2)

Recent anomalies

https://api-inference.bitdeer.ai/v1Aug 23, 2026, 5:39 AM

Self-claimed model: Qwen/Qwen3.8-2.4T-A95B
V2 predicted family: qwen
V3 best match: qwen/qwen3-max (sim 0.65)
Verdict: Full swap

https://api-inference.bitdeer.ai/v1Aug 23, 2026, 5:37 AM

Self-claimed model: Qwen/Qwen3.8-2.4T-A95B
V2 predicted family: qwen
V3 best match: qwen/qwen3-max (sim 0.64)
Verdict: Full swap

https://api.gnrt.dev/v1Aug 22, 2026, 7:15 PM

Self-claimed model: Qwen/Qwen3.8-2.4T-A95B
V2 predicted family: qwen
V3 best match: —
Verdict: Full swap

https://api.gnrt.dev/v1Aug 22, 2026, 4:13 PM

Self-claimed model: Qwen/Qwen3.8-2.4T-A95B
V2 predicted family: qwen
V3 best match: —
Verdict: Full swap

https://api.gnrt.dev/v1Aug 22, 2026, 3:02 PM

Self-claimed model: Qwen/Qwen3.8-2.4T-A95B
V2 predicted family: qwen
V3 best match: —
Verdict: Full swap

https://api.gnrt.dev/v1Aug 22, 2026, 2:45 PM

Self-claimed model: Qwen/Qwen3.8-2.4T-A95B
V2 predicted family: qwen
V3 best match: qwen/qwen3-max (sim 0.62)
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 qwen3824ta95b. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →