z-ai/glm-5.3 · Model identity and integrity report
Last updated：Aug 19, 2026, 11:34 AM · Data window: past 90 days
Measured results (90 days)
38 runs · 15 hosts · 36 anomalies
Family-fingerprint consistency
zhipu 30 ████████████████████░░░░░ 79% (suspicious) openai 3 ██░░░░░░░░░░░░░░░░░░░░░░░ 8% (suspicious) unknown 2 █░░░░░░░░░░░░░░░░░░░░░░░░ 5% (suspicious) deepseek 2 █░░░░░░░░░░░░░░░░░░░░░░░░ 5% (suspicious) xai 1 █░░░░░░░░░░░░░░░░░░░░░░░░ 3% (suspicious)
Observed V3 sub-model mapping
→ abstain 31 (82%) — low confidence → unknown 4 (11%) ⚠ mismatch → z-ai/glm-5 2 ( 5%) ⚠ mismatch sim=0.61 → deepseek/deepseek-v4-pro 1 ( 3%) ⚠ mismatch sim=0.66
Capability score distribution (auxiliary)
0-49 █████████████░░░░░░░░░░░░ 20 50-69 █████░░░░░░░░░░░░░░░░░░░░ 7 70-89 ███████░░░░░░░░░░░░░░░░░░ 11 90-100 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://api.a6api.com/v1 | 8 | 27 | 0% | — |
| https://api.amanai.dev/v1 | 5 | 26 | 0% | — |
| https://llm.goaichat.top/v1 | 4 | 34 | 0% | deepseek/deepseek-v4-pro (1) |
| https://api.inferhub.dev/v1 | 4 | 69 | 0% | — |
| https://api.cosmoshub.tech/v1 | 4 | 16 | 0% | z-ai/glm-5 (2) |
| https://rggwu8d.abc-tunnel.us/v1 | 2 | 0 | 0% | — |
| https://bandelbanget.xyz/v1 | 2 | 0 | 0% | — |
| https://router.lobakmerah.my.id/v1 | 2 | 75 | 0% | — |
| https://api.oxoapi.com/v1 | 1 | 9 | 0% | — |
| https://1.19848845.xyz/v1 | 1 | 0 | 0% | — |
| https://api.ngawi-proxy.cloud/v1 | 1 | 16 | 0% | — |
| https://wawapii.com/v1 | 1 | 72 | 0% | — |
| https://api.lmuai.com/v1 | 1 | 72 | 0% | — |
| https://pecut.aanserver.web.id/v1 | 1 | 71 | 0% | — |
| https://should-circular-beverages-attempting.trycloudflare.com/v1 | 1 | 70 | 0% | — |
Recent anomalies
https://api.oxoapi.com/v1Aug 19, 2026, 9:08 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- Family mismatch
https://llm.goaichat.top/v1Aug 19, 2026, 8:31 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.a6api.com/v1Aug 19, 2026, 8:17 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://rggwu8d.abc-tunnel.us/v1Aug 19, 2026, 5:10 AM
- Self-claimed model
- cbai/glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://rggwu8d.abc-tunnel.us/v1Aug 19, 2026, 5:05 AM
- Self-claimed model
- cbai/glm-5.3
- V2 predicted family
- xai
- V3 best match
- —
- Verdict
- Family mismatch
https://1.19848845.xyz/v1Aug 19, 2026, 4:47 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://llm.goaichat.top/v1Aug 19, 2026, 1:59 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- deepseek
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.a6api.com/v1Aug 19, 2026, 1:37 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.a6api.com/v1Aug 19, 2026, 1:27 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.a6api.com/v1Aug 18, 2026, 11:10 PM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.a6api.com/v1Aug 18, 2026, 11:03 PM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://llm.goaichat.top/v1Aug 18, 2026, 6:45 PM
- Self-claimed model
- glm-5.3
- V2 predicted family
- deepseek
- V3 best match
- deepseek/deepseek-v4-pro (sim 0.66)
- Verdict
- Full swap
https://wawapii.com/v1Aug 18, 2026, 5:17 PM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.inferhub.dev/v1Aug 18, 2026, 11:07 AM
- Self-claimed model
- zai/glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.inferhub.dev/v1Aug 18, 2026, 11:04 AM
- Self-claimed model
- cbcn/glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.cosmoshub.tech/v1Aug 18, 2026, 10:51 AM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.inferhub.dev/v1Aug 18, 2026, 10:23 AM
- Self-claimed model
- cbcn/glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://api.a6api.com/v1Aug 17, 2026, 2:45 PM
- Self-claimed model
- glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- Verdict
- Full swap
https://llm.goaichat.top/v1Aug 17, 2026, 12:14 PM
- Self-claimed model
- glm-5.3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.inferhub.dev/v1Aug 17, 2026, 9:10 AM
- Self-claimed model
- cbcn/glm-5.3
- V2 predicted family
- zhipu
- V3 best match
- —
- 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 z-ai/glm-5.3. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.