← Back to model

deepseekv32 · Model identity and integrity report

Last updated：Aug 31, 2026, 11:46 AM · Data window: past 90 days

Measured results (90 days)

4 runs · 4 hosts · 4 anomalies

Family-fingerprint consistency

  deepseek         4  █████████████████████████  100% (suspicious)

Observed V3 sub-model mapping

  → abstain                                  3 (75%) — low confidence
  → deepseek/deepseek-v3.2                   1 (25%) ⚠ mismatch sim=0.77

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://147ai.com/v1	1	0	0%	—
https://api2.68886868.xyz/v1	1	74	0%	—
https://api.ksm.moe/v1	1	66	0%	—
https://api.ikun.cloud-ip.cc/v1	1	72	0%	deepseek/deepseek-v3.2 (1)

Recent anomalies

https://147ai.com/v1Aug 31, 2026, 9:04 AM

Self-claimed model: deepseek-v3.2
V2 predicted family: deepseek
V3 best match: —
Verdict: V3 low confidence · Abstain

https://api2.68886868.xyz/v1Jul 31, 2026, 7:54 AM

Self-claimed model: [aws]deepseek-v3.2
V2 predicted family: deepseek
V3 best match: —
Verdict: V3 low confidence · Abstain

https://api.ksm.moe/v1Jul 30, 2026, 3:45 PM

Self-claimed model: deepseek-v3.2
V2 predicted family: deepseek
V3 best match: —
Verdict: Family mismatch

https://api.ikun.cloud-ip.cc/v1Jul 19, 2026, 11:55 AM

Self-claimed model: deepseek-v3.2
V2 predicted family: deepseek
V3 best match: deepseek/deepseek-v3.2 (sim 0.77)
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 deepseekv32. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.

Test an endpoint yourself →