openai/gpt-5.3-codex · Model identity and integrity report
Last updated：Jul 30, 2026, 11:59 AM · Data window: past 90 days
Verification capability — fingerprint reference available
- Family
- openai
- V3 sub-model reference
- Available ✓
- Checks
- 知識截止 · 能力探針 · 拒絕模板 · 拒絕溫度參數
- Family-fingerprint coverage
- 52 weighted rules · 8 feature categories
Concrete reference answers, refusal templates, and individual signal weights are not published because disclosure would help spoofing endpoints evade detection.
Measured results (90 days)
48 runs · 35 hosts · 43 anomalies
Family-fingerprint consistency
openai 39 ████████████████████░░░░░ 81% (expected) unknown 5 ███░░░░░░░░░░░░░░░░░░░░░░ 10% (suspicious) anthropic 4 ██░░░░░░░░░░░░░░░░░░░░░░░ 8% (suspicious)
Observed V3 sub-model mapping
→ openai/gpt-5.4 23 (48%) ⚠ mismatch sim=0.97 → abstain 18 (38%) — low confidence → unknown 6 (13%) ⚠ mismatch → openai/gpt-4.1 1 ( 2%) ⚠ mismatch sim=0.80
Capability score distribution (auxiliary)
0-49 ███████████░░░░░░░░░░░░░░ 20 50-69 ██░░░░░░░░░░░░░░░░░░░░░░░ 3 70-89 ███░░░░░░░░░░░░░░░░░░░░░░ 6 90-100 ██████████░░░░░░░░░░░░░░░ 19
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://www.moyu.info/v1 | 7 | 52 | 100% | openai/gpt-5.4 (2) |
| https://api.aigateways.cn/v1 | 3 | 95 | 100% | openai/gpt-5.4 (1) |
| https://api.quotaflow.ai/openai/v1 | 2 | 51 | 50% | — |
| https://yunwu.ai/v1 | 2 | 90 | 100% | openai/gpt-5.4 (2) |
| https://anyone.ai/v1 | 2 | 91 | 100% | openai/gpt-5.4 (2) |
| https://realmrouter.cn/v1 | 2 | 93 | 0% | openai/gpt-5.4 (2) |
| http://149.88.69.46:3001/v1 | 2 | 43 | 0% | — |
| https://sarang.id/v1 | 1 | 88 | 100% | openai/gpt-4.1 (1) |
| https://tokenapi.link/v1 | 1 | 88 | 100% | — |
| https://ciyuan.fast/v1 | 1 | 0 | 0% | — |
| https://code.shoestravel.xin/v1 | 1 | 5 | 0% | — |
| https://aikeypass.xyz/v1 | 1 | 0 | 0% | — |
| https://api.euzhi.com/v1 | 1 | 91 | 100% | — |
| https://api.fenno.ai/v1 | 1 | 92 | 100% | openai/gpt-5.4 (1) |
| https://codex.ciii.club/v1 | 1 | 0 | 100% | openai/gpt-5.4 (1) |
| https://4router.net/v1 | 1 | 92 | 100% | openai/gpt-5.4 (1) |
| https://codex-api-01.aiweber.com/v1 | 1 | 86 | 100% | openai/gpt-5.4 (1) |
| https://api.univibe.cc/openai/v1 | 1 | 0 | 100% | — |
| https://api.tangguo.xin/v1 | 1 | 0 | 100% | openai/gpt-5.4 (1) |
| https://www.daydreaming.work/v1 | 1 | 93 | 100% | openai/gpt-5.4 (1) |
| https://openai.chumeng.plus/v1 | 1 | 0 | 100% | openai/gpt-5.4 (1) |
| https://sub2api.itsfurry.com/v1 | 1 | 91 | 100% | — |
| https://www.aetherapi.cn/v1 | 1 | 0 | 100% | openai/gpt-5.4 (1) |
| https://tokenclub.top/v1 | 1 | 0 | 100% | openai/gpt-5.4 (1) |
| http://aaa.p8.ink/v1 | 1 | 89 | 100% | — |
Recent anomalies
https://sarang.id/v1Jul 15, 2026, 12:40 PM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-4.1 (sim 0.80)
- Verdict
- Sub-model downgrade
https://tokenapi.link/v1Jul 15, 2026, 3:48 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.quotaflow.ai/openai/v1Jun 22, 2026, 6:42 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.euzhi.com/v1Jun 13, 2026, 3:44 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.fenno.ai/v1Jun 1, 2026, 7:57 PM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 1.00)
- Verdict
- Sub-model downgrade
https://codex.ciii.club/v1Jun 1, 2026, 1:56 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.91)
- Verdict
- Sub-model downgrade
https://4router.net/v1May 29, 2026, 1:46 PM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.99)
- Verdict
- Sub-model downgrade
https://codex-api-01.aiweber.com/v1May 26, 2026, 10:25 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.91)
- Verdict
- Sub-model downgrade
https://www.moyu.info/v1May 26, 2026, 9:02 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://yunwu.ai/v1May 26, 2026, 8:32 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.94)
- Verdict
- Sub-model downgrade
https://www.moyu.info/v1May 26, 2026, 8:15 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://www.moyu.info/v1May 26, 2026, 8:08 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 1.00)
- Verdict
- Sub-model downgrade
https://yunwu.ai/v1May 26, 2026, 8:05 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.89)
- Verdict
- Sub-model downgrade
https://api.univibe.cc/openai/v1May 25, 2026, 7:03 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.tangguo.xin/v1May 24, 2026, 5:43 PM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.98)
- Verdict
- Sub-model downgrade
https://www.daydreaming.work/v1May 24, 2026, 7:02 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.98)
- Verdict
- Sub-model downgrade
https://www.moyu.info/v1May 22, 2026, 10:22 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.80)
- Verdict
- Sub-model downgrade
https://www.moyu.info/v1May 22, 2026, 6:56 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://openai.chumeng.plus/v1May 21, 2026, 11:13 AM
- Self-claimed model
- gpt-5.3-codex
- V2 predicted family
- openai
- V3 best match
- openai/gpt-5.4 (sim 0.98)
- Verdict
- Sub-model downgrade
https://sub2api.itsfurry.com/v1May 21, 2026, 10:42 AM
- Self-claimed model
- gpt-5.3-codex
- 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 openai/gpt-5.3-codex. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.