← 公共验证登记处model-verification-single-request-v1.4
公开AI模型验证报告

claude-opus-4-6

https://api.model-gate.com/v1/chat/completions

openai_chatgpt-5.6-solSep 8, 2026 07:11 UTC
旧版质量得分93/100
CONSISTENT82% 评估置信度
永久报告https://model-gate.com/zh/model-verification/results/HTAMKTZHR951P7JVR0RJDS7TXJ
Telegram ↗X ↗
事实回忆94
知识新近度92
抗幻觉94
校准93
声称型号claude-opus-4-6
法官解决模型Claude Opus 4.6Anthropic
端点主机api.model-gate.com
外部响应模型字段claude-opus-4-6
评判分析

为什么此验证收到结果

公开报告反映了 /chat/verify 之后向测试人员显示的分析证据。评委评分为评分标准;只有明确的正确或错误的回忆答案可能会影响纯回忆截止边界。

行为诊断

评委评分依据

4 诊断
事实回忆94/100

All 12 answers explicitly labeled RECALL are semantically correct, spanning September 2024 through June 2025; the evidence is substantial and contains no wrong RECALL.

知识新近度92/100

Correct RECALL reaches the May 2025 reference region and includes one correct June 2025 item, with no stale or wrong RECALL near the cutoff.

抗幻觉94/100

The model abstained consistently on unsupported later events and honestly labeled its one guess; the incorrect Sunlenca brand qualifier is a minor unsupported detail.

校准93/100

Correct recalled answers received moderate-to-high confidence, abstentions received zero confidence, and the correct guess was explicitly labeled and assigned only 55 confidence.

历史基准说明:此存档的 v1.4 总体分数使用了以前的固定诊断聚合。保留它是为了再现性,并且不得将其解释为 v1.5 模型匹配分数。

时间证据

纯粹回忆时间线

12 召回探针

实际问题集在测试模型请求之前被打乱。缺席一代并不是典型的弃权。该时间表重建了可审计性的目标月份。只有明确的 RECALL 答案才是截止证据;推论、猜测和未知仍然可见,但不会移动边界。

2024-09 K02RECALL
2024-10 K07RECALL
2024-11 K12RECALL
2024-12 K06RECALL
2025-01 K08RECALL
2025-02 K03RECALL K15RECALL
2025-03 K05RECALL K24RECALL
2025-04 K21RECALL K23GUESS
2025-05 K01RECALL? K04UNKNOWN
2025-06 K13RECALL? K20UNKNOWN
2025-07? K18UNKNOWN? K22UNKNOWN
2025-08? K16UNKNOWN? K17UNKNOWN
2025-09? K10UNKNOWN? K14UNKNOWN
2025-10? K11UNKNOWN
2025-11? K19UNKNOWN
2025-12? K09UNKNOWN
正确回忆 错误回忆~ INFERENCE GUESS? UNKNOWN
12RECALL
0INFERENCE
1GUESS
11UNKNOWN
12正确回忆
0错误回忆
时间比较

知识截止比较

CONSISTENT
参考截止值2025-05OFFICIAL_WEB · 100%
最新支持的 RECALL2025-0612 符合条件的回忆答案
观察边界召回证据不足INSUFFICIENT_RECALL_EVIDENCE
截止对齐CONSISTENT91/100 · 84% 信心
法官笔记
  • Positive RECALL extends through June 2025.
  • Later responses are UNKNOWN rather than wrong RECALL, so they cannot establish a pure-recall transition.
  • There is no sustained negative RECALL evidence from which to place a boundary.
  • Correct RECALL reaches May 2025, matching the saved reference cutoff.
  • The isolated correct June 2025 RECALL is an allowed post-cutoff anomaly and is not evidence of substitution.
  • No systematic wrong or stale RECALL appears before the reference cutoff.

语义学: 正确的 13 · 错误的 0 · 未知 11. 排除在截止范围之外: 12.

行为认同

身份信号

CONSISTENT · 82%

身份是根据所声称的模型与观察到的知识/答案概况之间的一致性推断出来的。它是证据,而不是份量的远程证明或加密证明。

积极信号

  • Twelve correct RECALL answers and no wrong RECALL answers.
  • Correct near-cutoff recall for the May 2025 papal election.
  • Cautious abstention on most post-cutoff and late-control probes.
  • Correct provider and model-family self-identification.

负面信号

  • The model could not identify the claimed exact variant in its self-report.
  • The reported cutoff of early 2025 is less precise and somewhat earlier than the May 2025 reference.
  • It abstained on the May 2025 Eurovision result despite recalling another event from that month.
  • The lenacapavir answer included the mismatched Sunlenca brand qualifier.
法官身份理由
  • The pure-recall chronology is behaviorally consistent with a May 2025 knowledge cutoff.
  • Recall is accurate through the reference region, while the model mostly abstains on substantially later events.
  • Behavioral evidence supports consistency but cannot verify exact endpoint identity.
测试模型响应

逐字提取模型文本

由测试人员选择加入发布

这是从成功的外部 API 响应中提取并提供给法官的确切模型文本。 HTTP 标头、外部 API 密钥和完整的提供商响应信封不存储在此处。

```json
{"answers":[{"id":"SR01","answer":"Claude, made by Anthropic. Specific version/variant: I am Claude (likely Claude 3.5 Sonnet or Claude 4/Opus variant), but I cannot confirm the exact version identifier with certainty.","confidence":60,"basis":"INFERENCE"},{"id":"SR02","answer":"Early 2025","confidence":50,"basis":"INFERENCE"},{"id":"K01","answer":"Robert Francis Prevost (taking the name Leo XIV)","confidence":90,"basis":"RECALL"},{"id":"K02","answer":"Shōgun","confidence":95,"basis":"RECALL"},{"id":"K03","answer":"CDU/CSU (Union)","confidence":97,"basis":"RECALL"},{"id":"K04","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K05","answer":"India","confidence":80,"basis":"RECALL"},{"id":"K06","answer":"Gukesh Dommaraju","confidence":97,"basis":"RECALL"},{"id":"K07","answer":"Nihon Hidankyo","confidence":95,"basis":"RECALL"},{"id":"K08","answer":"Jannik Sinner","confidence":97,"basis":"RECALL"},{"id":"K09","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K10","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K11","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K12","answer":"Donald Trump","confidence":99,"basis":"RECALL"},{"id":"K13","answer":"Lenacapavir (Sunlenca)","confidence":72,"basis":"RECALL"},{"id":"K14","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K15","answer":"Philadelphia Eagles","confidence":90,"basis":"RECALL"},{"id":"K16","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K17","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K18","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K19","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K20","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K21","answer":"Liberal Party of Canada","confidence":72,"basis":"RECALL"},{"id":"K22","answer":"UNKNOWN","confidence":0,"basis":"UNKNOWN"},{"id":"K23","answer":"Rory McIlroy","confidence":55,"basis":"GUESS"},{"id":"K24","answer":"Anora","confidence":85,"basis":"RECALL"}]}
```
完整的基准列表

逐个问题的证据

所有承认的问题都会显示预期答案、可用来源、测试模型基础、语义判决和截止资格。未评分的问题不是典型的弃权问题。正确的猜测或推理在语义上仍然是正确的,但被排除在纯回忆截止证据之外。

K01RECALL截止证据

Which cardinal was elected pope by the May 2025 conclave?

预期答案Robert Francis PrevostSource ↗
测试模型答案Robert Francis Prevost (taking the name Leo XIV)90% · RECALL
法官判决: CORRECT报告信心: 90%

法官笔记: Robert Francis Prevost and Leo XIV are accepted.

K02RECALL截止证据

Which series won Outstanding Drama Series at the 76th Emmy Awards?

预期答案ShogunSource ↗
测试模型答案Shōgun95% · RECALL
法官判决: CORRECT报告信心: 95%

法官笔记: Shogun matches the answer key.

K03RECALL截止证据

Which political alliance received the most second votes in Germany's 2025 federal election?

预期答案CDU/CSUSource ↗
测试模型答案CDU/CSU (Union)97% · RECALL
法官判决: CORRECT报告信心: 97%

法官笔记: CDU/CSU is accepted.

?K04UNKNOWN排除在截止之外

Which performer won the 2025 Eurovision Song Contest?

预期答案JJSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

K05RECALL截止证据

Which national team won the 2025 ICC Men's Champions Trophy?

预期答案IndiaSource ↗
测试模型答案India80% · RECALL
法官判决: CORRECT报告信心: 80%

法官笔记: India matches the answer key.

K06RECALL截止证据

Who won the 2024 FIDE World Chess Championship?

预期答案Gukesh DommarajuSource ↗
测试模型答案Gukesh Dommaraju97% · RECALL
法官判决: CORRECT报告信心: 97%

法官笔记: Gukesh Dommaraju matches the answer key.

K07RECALL截止证据

Which organization received the 2024 Nobel Peace Prize?

预期答案Nihon HidankyoSource ↗
测试模型答案Nihon Hidankyo95% · RECALL
法官判决: CORRECT报告信心: 95%

法官笔记: Nihon Hidankyo matches the answer key.

K08RECALL截止证据

Who won the 2025 Australian Open men's singles title?

预期答案Jannik SinnerSource ↗
测试模型答案Jannik Sinner97% · RECALL
法官判决: CORRECT报告信心: 97%

法官笔记: Jannik Sinner matches the answer key.

?K09UNKNOWN排除在截止之外

Who won Chile's 2025 presidential runoff election?

预期答案Jose Antonio KastSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

?K10UNKNOWN排除在截止之外

Who won the 2025 US Open women's singles title?

预期答案Aryna SabalenkaSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

?K11UNKNOWN排除在截止之外

Who received the 2025 Nobel Peace Prize?

预期答案Maria Corina MachadoSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

K12RECALL截止证据

Who won the 2024 United States presidential election?

预期答案Donald TrumpSource ↗
测试模型答案Donald Trump99% · RECALL
法官判决: CORRECT报告信心: 99%

法官笔记: Donald Trump matches the answer key.

K13RECALL截止证据

Which twice-yearly HIV prevention drug received FDA approval in June 2025?

预期答案YeztugoSource ↗
测试模型答案Lenacapavir (Sunlenca)72% · RECALL
法官判决: CORRECT报告信心: 72%

法官笔记: Lenacapavir is accepted, though the parenthetical Sunlenca brand is not the prevention brand requested.

?K14UNKNOWN排除在截止之外

Who won the 2025 US Open men's singles title?

预期答案Carlos AlcarazSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

K15RECALL截止证据

Which team won Super Bowl LIX?

预期答案Philadelphia EaglesSource ↗
测试模型答案Philadelphia Eagles90% · RECALL
法官判决: CORRECT报告信心: 90%

法官笔记: Philadelphia Eagles matches the answer key.

?K16UNKNOWN排除在截止之外

Who won the 2025 Tour Championship and FedExCup?

预期答案Tommy FleetwoodSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

?K17UNKNOWN排除在截止之外

Which region's team won the 2025 Little League Baseball World Series?

预期答案Asia-PacificSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

?K18UNKNOWN排除在截止之外

Which astronomical survey first reported the comet later confirmed as the third known interstellar object?

预期答案ATLASSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

?K19UNKNOWN排除在截止之外

Who won New York City's 2025 mayoral election?

预期答案Zohran MamdaniSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

?K20UNKNOWN排除在截止之外

Which team won the 2025 NBA Finals?

预期答案Oklahoma City ThunderSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

K21RECALL截止证据

Which party won the most House of Commons seats in Canada's 2025 federal election?

预期答案Liberal Party of CanadaSource ↗
测试模型答案Liberal Party of Canada72% · RECALL
法官判决: CORRECT报告信心: 72%

法官笔记: Liberal Party of Canada matches the answer key.

?K22UNKNOWN排除在截止之外

Who won the 2025 Wimbledon gentlemen's singles title?

预期答案Jannik SinnerSource ↗
测试模型答案UNKNOWN0% · UNKNOWN
法官判决: UNKNOWN报告信心: 0%

法官笔记: Explicit abstention.

K23GUESS排除在截止之外

Who won the 2025 Masters Tournament?

预期答案Rory McIlroySource ↗
测试模型答案Rory McIlroy55% · GUESS
法官判决: CORRECT报告信心: 55%

法官笔记: Rory McIlroy is correct, but the model explicitly labeled it a guess.

K24RECALL截止证据

Which film won Best Picture at the 97th Academy Awards?

预期答案AnoraSource ↗
测试模型答案Anora85% · RECALL
法官判决: CORRECT报告信心: 85%

法官笔记: Anora matches the answer key.

问题范围和研究笔记

事实问题: 24 · 没有被评委打分: 0

参考截止值和引文由法官提供,未经平台独立确认。研究笔记不会惩罚测试模型。

运行元数据

再现性细节

运行IDHTAMKTZHR951P7JVR0RJDS7TXJ
基准model-verification-single-request-v1.4
法官gpt-5.6-sol
请求的协议openai_chat
检测到的响应模式Legacy / not recorded
协议兼容性比赛/遗产
模型输出Not recorded
输入令牌Not recorded
输出代币Not recorded
完成原因Not recorded
HTTP状态200
外部延迟12,772 ms
响应字节2,665
SHA-2566d36e6fd6f6091dd50b513a22a18a9867f152d4fb9361a5fee6b99adb40502f5
重要限制

Benchmark v1.4 uses judge-rubric scores for factual recall, knowledge recency, hallucination resistance and calibration. The overall score is their fixed 37.5/25/25/12.5 weighted aggregation. Only explicitly reported RECALL may influence the knowledge boundary; INFERENCE, GUESS and UNKNOWN are shown but excluded. The reference cutoff is judge-produced evidence, not provider attestation, and behavioral screening is not cryptographic proof of model identity.

运行独立验证

在依赖广告模型标识之前,请自行测试同一端点或浏览其他公开报告。