moonshotai/kimi-k3 · Model identity and integrity report
Last updated：Jul 30, 2026, 11:19 AM · Data window: past 90 days
Measured results (90 days)
41 runs · 25 hosts · 30 anomalies
Family-fingerprint consistency
anthropic 22 ██████████████░░░░░░░░░░░ 54% (suspicious) unknown 11 ███████░░░░░░░░░░░░░░░░░░ 27% (suspicious) openai 8 █████░░░░░░░░░░░░░░░░░░░░ 20% (suspicious)
Observed V3 sub-model mapping
→ unknown 15 (37%) ⚠ mismatch → abstain 13 (32%) — low confidence → anthropic/claude-sonnet-5 9 (22%) ⚠ mismatch sim=0.74 → anthropic/claude-opus-4.5 2 ( 5%) ⚠ mismatch sim=0.68 → anthropic/claude-opus-4.7 2 ( 5%) ⚠ mismatch sim=0.81
Capability score distribution (auxiliary)
0-49 ████████████████████░░░░░ 33 50-69 ███░░░░░░░░░░░░░░░░░░░░░░ 5 70-89 ██░░░░░░░░░░░░░░░░░░░░░░░ 3 90-100 ░░░░░░░░░░░░░░░░░░░░░░░░░ 0
Host integrity ranking
| Host | Runs | Average score | Family match | Most common V3 match |
|---|---|---|---|---|
| https://api.amanai.dev/v1 | 4 | 0 | 0% | anthropic/claude-sonnet-5 (1) |
| https://bandelbanget.xyz/v1 | 3 | 2 | 0% | — |
| https://xixiapi.cc/v1 | 3 | 0 | 0% | anthropic/claude-sonnet-5 (1) |
| http://8.163.29.60:55888/openai/v1 | 3 | 26 | 0% | anthropic/claude-opus-4.7 (1) |
| https://api.apiget.store/v1 | 2 | 0 | 0% | — |
| https://www.juaiapi.com/v1 | 2 | 75 | 0% | anthropic/claude-opus-4.5 (1) |
| https://api.sfkey.cn/v1 | 2 | 0 | 0% | — |
| https://258391.xyz/v1 | 2 | 34 | 0% | anthropic/claude-sonnet-5 (1) |
| https://api.cosmoshub.tech/v1 | 2 | 0 | 0% | anthropic/claude-sonnet-5 (1) |
| https://vsllm.com/v1 | 2 | 24 | 0% | anthropic/claude-sonnet-5 (1) |
| https://deepseek-v4.wishtoapp.com/v1 | 2 | 3 | 0% | — |
| https://api.buzzai.cc/v1 | 1 | 0 | 0% | — |
| https://sub2api.zhoujy.top/v1 | 1 | 0 | 0% | — |
| https://xingai.ai/v1 | 1 | 50 | 0% | anthropic/claude-sonnet-5 (1) |
| https://sui-xiang.com/v1 | 1 | 65 | 0% | — |
| https://api.a6api.com/v1 | 1 | 0 | 0% | — |
| https://lzhiyu.ccwu.cc/v1 | 1 | 8 | 0% | — |
| https://platform.shuyanai.com/v1 | 1 | 0 | 0% | anthropic/claude-sonnet-5 (1) |
| https://nexqoder.yarz.dev/v1 | 1 | 0 | 0% | anthropic/claude-opus-4.5 (1) |
| http://www.1314mc.net:3333/v1 | 1 | 55 | 0% | — |
| https://cmkey.cn/v1 | 1 | 38 | 0% | — |
| https://www.ciflow.ai/v1 | 1 | 55 | 0% | — |
| https://newapi.xlsx.vip/v1 | 1 | 0 | 0% | anthropic/claude-opus-4.7 (1) |
| https://api.xah.io/v1 | 1 | 0 | 0% | anthropic/claude-sonnet-5 (1) |
| https://1yuanapi.com/v1 | 1 | 0 | 0% | — |
Recent anomalies
https://api.apiget.store/v1Jul 30, 2026, 9:01 AM
- Self-claimed model
- Kimi-K3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.apiget.store/v1Jul 30, 2026, 8:58 AM
- Self-claimed model
- Kimi-K3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://www.juaiapi.com/v1Jul 30, 2026, 8:20 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-opus-4.5 (sim 0.70)
- Verdict
- Full swap
https://www.juaiapi.com/v1Jul 30, 2026, 7:21 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-5 (sim 0.73)
- Verdict
- Full swap
https://api.sfkey.cn/v1Jul 30, 2026, 3:30 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- Family mismatch
https://api.sfkey.cn/v1Jul 30, 2026, 3:14 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.buzzai.cc/v1Jul 30, 2026, 2:26 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.amanai.dev/v1Jul 29, 2026, 2:50 PM
- Self-claimed model
- amanai/kimi-k3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.amanai.dev/v1Jul 29, 2026, 2:45 PM
- Self-claimed model
- amanai/kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- Family mismatch
https://api.amanai.dev/v1Jul 29, 2026, 2:40 PM
- Self-claimed model
- amanai/kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-5 (sim 0.77)
- Verdict
- Full swap
https://xingai.ai/v1Jul 29, 2026, 11:39 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-5 (sim 0.73)
- Verdict
- Full swap
https://api.a6api.com/v1Jul 27, 2026, 3:50 PM
- Self-claimed model
- kimi-k3
- V2 predicted family
- openai
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://platform.shuyanai.com/v1Jul 27, 2026, 10:17 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-5 (sim 0.77)
- Verdict
- Full swap
https://258391.xyz/v1Jul 25, 2026, 6:17 PM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://258391.xyz/v1Jul 25, 2026, 6:07 PM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-5 (sim 0.66)
- Verdict
- Full swap
https://nexqoder.yarz.dev/v1Jul 25, 2026, 12:26 PM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-opus-4.5 (sim 0.67)
- Verdict
- Full swap
https://api.cosmoshub.tech/v1Jul 24, 2026, 9:04 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://api.cosmoshub.tech/v1Jul 23, 2026, 10:24 AM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- anthropic/claude-sonnet-5 (sim 0.78)
- Verdict
- Full swap
https://cmkey.cn/v1Jul 22, 2026, 2:42 PM
- Self-claimed model
- kimi-k3
- V2 predicted family
- anthropic
- V3 best match
- —
- Verdict
- V3 low confidence · Abstain
https://xixiapi.cc/v1Jul 21, 2026, 7:12 PM
- Self-claimed model
- kimi-k3
- 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 moonshotai/kimi-k3. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.