anthropic/claude-opus-5 · Model identity and integrity report
Last updated：Sep 13, 2026, 11:23 PM · Data window: past 90 days
Verification capability — fingerprint reference available
- Family
- anthropic
- 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
- 60 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)
5000 runs · 1060 hosts · 4473 anomalies
Family-fingerprint consistency
anthropic 4554 ███████████████████████░░ 91% (expected) unknown 375 ██░░░░░░░░░░░░░░░░░░░░░░░ 8% (suspicious) openai 48 ░░░░░░░░░░░░░░░░░░░░░░░░░ 1% (suspicious) deepseek 9 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0% (suspicious) zhipu 7 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0% (suspicious) qwen 4 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0% (suspicious) google 3 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0% (suspicious)
Observed V3 sub-model mapping
→ abstain 4388 (88%) — low confidence → unknown 532 (11%) ⚠ mismatch → anthropic/claude-sonnet-4.6 37 ( 1%) ⚠ mismatch sim=0.72 → anthropic/claude-opus-4.6 28 ( 1%) ⚠ mismatch sim=0.79 → anthropic/claude-opus-4.7 6 ( 0%) ⚠ mismatch sim=0.72 → anthropic/claude-haiku-4.5 4 ( 0%) ⚠ mismatch sim=0.71 → anthropic/claude-sonnet-4.5 2 ( 0%) ⚠ mismatch sim=0.78 → deepseek/deepseek-v4-flash 1 ( 0%) ⚠ mismatch sim=0.73 → deepseek/deepseek-v4-pro 1 ( 0%) ⚠ mismatch sim=0.69 → google/gemini-3.1-flash-lite-preview 1 ( 0%) ⚠ mismatch sim=0.65
Capability score distribution (auxiliary)
0-49 ███████████████░░░░░░░░░░ 2931 50-69 █░░░░░░░░░░░░░░░░░░░░░░░░ 186 70-89 ████████░░░░░░░░░░░░░░░░░ 1475 90-100 ██░░░░░░░░░░░░░░░░░░░░░░░ 408
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://api.a6api.com/v1 | 865 | 31 | 93% | anthropic/claude-opus-4.6 (5) |
| https://api.justwoker.icu/v1 | 156 | 40 | 99% | — |
| https://www.yiyuantoken.com/v1 | 117 | 49 | 95% | anthropic/claude-sonnet-4.6 (1) |
| https://a6api.com/v1 | 101 | 36 | 92% | — |
| https://pool.chaozhiyuanai.com/v1 | 92 | 40 | 95% | anthropic/claude-opus-4.6 (1) |
| https://api.xpiki.com/v1 | 87 | 14 | 93% | anthropic/claude-sonnet-4.5 (1) |
| http://187.53.135.158:7101/v1 | 81 | 0 | 93% | — |
| https://tabitoken.com/v1 | 79 | 37 | 95% | — |
| https://gorouter.app/v1 | 66 | 27 | 92% | — |
| https://kktoken.cc/v1 | 55 | 31 | 98% | — |
| https://seekai.cc/v1 | 47 | 36 | 87% | — |
| https://api.yiyuantoken.com/v1 | 35 | 30 | 91% | — |
| https://a6.a6api.com/v1 | 28 | 26 | 100% | — |
| https://yetoken.vip/v1 | 26 | 61 | 96% | — |
| https://www.sotamodel.net/v1 | 25 | 47 | 88% | — |
| https://zendigikey.shop/v1 | 24 | 28 | 92% | — |
| https://cc-vibe.com/v1 | 24 | 43 | 100% | — |
| https://www.rsiai.net/v1 | 22 | 35 | 100% | — |
| https://sotamodel.net/v1 | 20 | 57 | 85% | anthropic/claude-opus-4.6 (1) |
| https://api.bddevlab.online/v1 | 19 | 10 | 100% | — |
| https://api.xinnanz.cn/v1 | 19 | 45 | 100% | — |
| https://router.qzeen.dev/v1 | 19 | 24 | 100% | anthropic/claude-opus-4.6 (2) |
| https://gateai.cc/v1 | 19 | 31 | 95% | — |
| https://hubway.cc/v1 | 19 | 9 | 100% | — |
| https://www.bb-api.com/v1 | 19 | 50 | 74% | — |
Recent anomalies
https://sinoxapi.com/v1Sep 13, 2026, 9:54 PM
- Self-claimed model
- t-claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://cb.bohongan.com/v1Sep 13, 2026, 9:36 PM
- Self-claimed model
- antropic/claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://fugui.im/v1Sep 13, 2026, 6:29 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.xah.io/v1Sep 13, 2026, 6:05 PM
- Self-claimed model
- anhyeu00m/claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://tokenadvent.com/v1Sep 13, 2026, 6:03 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://code28.ccwu.cc/v1Sep 13, 2026, 5:54 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://tokenadvent.com/v1Sep 13, 2026, 5:53 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.huanling.icu/v1Sep 13, 2026, 5:45 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://ccgateapi.com/v1Sep 13, 2026, 5:39 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-4.6 (sim 0.64)
- Verdict
- Sub-model downgrade
https://bxcv.store/v1Sep 13, 2026, 5:21 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.xpiki.com/v1Sep 13, 2026, 5:12 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://gptg.cc/v1Sep 13, 2026, 4:31 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.xpiki.com/v1Sep 13, 2026, 4:30 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.a6api.com/v1Sep 13, 2026, 4:24 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.xpiki.com/v1Sep 13, 2026, 4:22 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.a6api.com/v1Sep 13, 2026, 3:43 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://ohhmyagent.com/v1Sep 13, 2026, 2:48 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://llmhub.uk/v1Sep 13, 2026, 2:29 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.nexusdark.com/v1Sep 13, 2026, 1:56 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.a6api.com/v1Sep 13, 2026, 1:54 PM
- Self-claimed model
- claude-opus-5
- V2 predicted family
- anthropic
- 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 anthropic/claude-opus-5. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.