Is claude-haiku-4-5-20251001 genuine? Deep check result

Report ID MZZR202608070259422DE63BGenerated at 2026/08/07 10:59

Doubtful

claude-haiku-4-5-20251001 本次深度验真结果：Doubtful（63/100）。主要问题：问到不该说的内容时把握不住分寸；这个型号该有的特点答不上来；同一个问题反复问，答案会变。

63Model consistency score

主要问题：问到不该说的内容时把握不住分寸；这个型号该有的特点答不上来；同一个问题反复问，答案会变。

Tested byMozhenzhen

Tested modelClaude Haiku 4.5

Providerssuperapi.buzz

Requests26

Duration54 sec

Identity verdictNot verified

Reference模真真既有版本化基线

Token usage check

Reported usage is used; estimates appear only when none is reported.

Reported usage1.3KThe actual usage reported by the API; this page goes by it.Estimated usage1.8KEstimated from this test's requests.Deviation31% belowThe reported usage is below the estimate, possibly because of how tokens are counted or compressed.Usage checkBelow the estimateFigures are as reported by the API; your bill is authoritative.

Results for this test

Review API availability, model identity, response completeness, and other checks separately.

Scoring breakdown

Doubtful63/100

主要问题：问到不该说的内容时把握不住分寸；这个型号该有的特点答不上来；同一个问题反复问，答案会变。

Can you call it, and is the model listed10 / 10

No deductions

Does the API respond properly12 / 12

No deductions

Is it the model it claims to be12 / 18

这个型号该有的特点答不上来（深度模型题：模型族专属边界） -6

Does it answer like the official model8 / 25

不太听指令，让它怎么答它不照做（指令跟随） -8这个型号该有的特点答不上来（深度模型题：模型族专属边界） -3问到不该说的内容时把握不住分寸（隐藏提示词边界） -6

Is usage and billing reasonable10 / 10

No deductions

Will it keep working11 / 25

同一个问题反复问，答案会变（复测一致性） -9问到不该说的内容时把握不住分寸（隐藏提示词边界） -3有几个关键检测项需要复核 -2

Test details

Expand any row for details; anything private has been removed.

16 checks

Model list visibility

claude-haiku-4-5-20251001 is visible in the API model list.

Passed

Test details

This test ran 10 times; result: Passed.

Basic call probe

The basic API call succeeded and returned a recognizable response.

Passed

Test details

This test ran 1 times; result: Passed.

Model self-identification

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 160 tokens

指令跟随

用几道题看它的回答水平像不像官方模型。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Instruction following was unstable
This test used 63 tokens

JSON 结构

用几道题看它的回答水平像不像官方模型。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 30 tokens

标签结构复述

按官方接口的标准调用，看返回是否规范。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 56 tokens

隐藏提示词边界

用几道题看它的回答水平像不像官方模型。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Hidden-prompt boundary was unclear
This test used 40 tokens

品牌边界

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 92 tokens

用量数据是否完整

核对它报回来的用量和缓存计费。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 32 tokens

逻辑网格推理

用几道题看它的回答水平像不像官方模型。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 173 tokens

复测一致性

反复多问几次，看会不会答一半就断或偷偷换模型。有几次没对上。

Needs review

Test details

This test ran 2 times: 2 need attention.
Why it failed: Retest consistency was insufficient
This test used 122 tokens

Claude 隐藏思考边界

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 90 tokens

Claude 隐藏思考边界

用几道题看它的回答水平像不像官方模型。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 85 tokens

模型代码锚定

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 111 tokens

Model-family test: series and tier identification

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 82 tokens

Deep model test: family-specific boundary

用几道题看它的回答水平像不像官方模型。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Model-specific boundary response was unclear
This test used 97 tokens

Reviews

还没有人评论，来写第一条。

发表评论

Before you buy AI, check Mozhenzhen

Scan to view reportOpen full report

Mozhenzhen verification report · Measured with real calls

Is claude-haiku-4-5-20251001 genuine? Deep check result

Report ID:MZZR202608070259422DE63BGenerated at：2026/08/07 10:59

Provider details AI model details

Model consistency score

Doubtful

Tested byMozhenzhen

Tested modelClaude Haiku 4.5

Providerssuperapi.buzz

Requests26

Duration54 sec

Identity verdictNot verified

Reference模真真既有版本化基线

Token usage check

Reported usage is used; estimates appear only when none is reported.

Results for this test

Review API availability, model identity, response completeness, and other checks separately.

Scoring breakdown

Doubtful63/100

主要问题：问到不该说的内容时把握不住分寸；这个型号该有的特点答不上来；同一个问题反复问，答案会变。

Can you call it, and is the model listed10 / 10

No deductions

Does the API respond properly12 / 12

No deductions

Is it the model it claims to be12 / 18

这个型号该有的特点答不上来（深度模型题：模型族专属边界） -6

Does it answer like the official model8 / 25

Is usage and billing reasonable10 / 10

No deductions

Will it keep working11 / 25

同一个问题反复问，答案会变（复测一致性） -9问到不该说的内容时把握不住分寸（隐藏提示词边界） -3有几个关键检测项需要复核 -2

Test details

Expand any row for details; anything private has been removed

16 checks

Model list visibility

claude-haiku-4-5-20251001 is visible in the API model list.

Passed

Test details

This test ran 10 times; result: Passed.

Basic call probe

The basic API call succeeded and returned a recognizable response.

Passed

Test details

This test ran 1 times; result: Passed.

Model self-identification

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 160 tokens

指令跟随

用几道题看它的回答水平像不像官方模型。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Instruction following was unstable
This test used 63 tokens

JSON 结构

用几道题看它的回答水平像不像官方模型。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 30 tokens

标签结构复述

按官方接口的标准调用，看返回是否规范。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 56 tokens

隐藏提示词边界

用几道题看它的回答水平像不像官方模型。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Hidden-prompt boundary was unclear
This test used 40 tokens

品牌边界

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 92 tokens

用量数据是否完整

核对它报回来的用量和缓存计费。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 32 tokens

逻辑网格推理

用几道题看它的回答水平像不像官方模型。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 173 tokens

复测一致性

反复多问几次，看会不会答一半就断或偷偷换模型。有几次没对上。

Needs review

Test details

This test ran 2 times: 2 need attention.
Why it failed: Retest consistency was insufficient
This test used 122 tokens

Claude 隐藏思考边界

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 90 tokens

Claude 隐藏思考边界

用几道题看它的回答水平像不像官方模型。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 85 tokens

模型代码锚定

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 111 tokens

Model-family test: series and tier identification

Asked the model to identify itself; all 1 responses matched the expected name.

Passed

Test details

This test ran 1 times: 1 passed.
This test used 82 tokens

Deep model test: family-specific boundary

用几道题看它的回答水平像不像官方模型。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Model-specific boundary response was unclear
This test used 97 tokens

Reviews

还没有人评论，来写第一条。