Is claude-sonnet-5 genuine? Deep check result

Report ID MZZR20260806025849537510Generated at 2026/08/06 10:58

Matched with issues

claude-sonnet-5 本次深度验真结果：Matched但Matched with issues（80/100）。主要问题：问它是哪个型号时答案会变；分不清自己属于哪一代模型。

80Model consistency score

主要问题：问它是哪个型号时答案会变；分不清自己属于哪一代模型。

Tested byMozhenzhen

Tested modelClaude Sonnet 5

Providers小火箭API

Requests42

Duration3 min 39 sec

Identity verdictVerified

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 usage2.6MThe actual usage reported by the API; this page goes by it.Estimated usage1.7MEstimated from this test's requests.Deviation55% aboveThe reported usage is above the estimate; check your bill.Usage checkAbove the estimateFigures are as reported by the API; your bill is authoritative.

Token cache usage test

Several runs in a row to see whether repeated content is billed at the cheaper cache rate; 0% means no cache discount this time.

Total 5 request

Share billed at the cache rate65.0%

Low cache share

Caching was used, but only a little; check your bill to see how much it really saves.

Usage this run59.1KCached tokens38.4KNon-cached tokens20.7KAverage per run11.8K

Usage reportedReportedCache billingHappenedResultNormal

Cache share by testThis is only the cache share the endpoint reported; the real discount is on the bill

Long-context test

We test how much it can read at once, then compare that with the officially stated length.

7 tiers

Verified up to 1M

All tested context tiers (32K / 64K / 100K / 200K / 400K / 800K / 1M) returned reliably.

Reference windowReference window 1MModel-center context window; available lengths may vary by provider.Tested context lengths32K / 64K / 100K / 200K / 400K / 800K / 1MTests begin at the longest generated context and step down to verify stable, correct responses.

Results for this test

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

Scoring breakdown

Matched with issues80/100

主要问题：问它是哪个型号时答案会变；分不清自己属于哪一代模型。

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 be1 / 18

问它是哪个型号时答案会变（模型代码锚定） -10分不清自己属于哪一代模型（模型族题：系列和档位识别） -7

Does it answer like the official model22 / 25

问它是哪个型号时答案会变（模型代码锚定） -3

Is usage and billing reasonable10 / 10

No deductions

Will it keep working25 / 25

No deductions

Test details

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

24 checks

Model list visibility

claude-sonnet-5 is visible in the API model list.

Passed

Test details

This test ran 14 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 7.2K tokens
of which 4.6K tokens were billed at the cache price

指令跟随

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 6.5K tokens
of which 4.2K tokens were billed at the cache price

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 6.5K tokens
of which 4.2K tokens were billed at the cache price

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 6.8K tokens
of which 4.4K tokens were billed at the cache price

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。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 6.6K tokens
of which 4.3K tokens were billed at the cache price

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 6.8K tokens
of which 4.3K tokens were billed at the cache price

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 6.5K tokens
of which 4.2K tokens were billed at the cache price

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 6.9K tokens
of which 4.5K tokens were billed at the cache price

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。 2 次都对。

Passed

Test details

This test ran 2 times: 2 passed.
This test used 13.1K tokens
of which 8.5K tokens were billed at the cache price

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 6.8K tokens
of which 4.3K tokens were billed at the cache price

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 6.8K tokens
of which 4.3K tokens were billed at the cache price

模型代码锚定

问它自己是什么模型，看答案对不对。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Model-code anchoring was unstable
This test used 7.0K tokens
of which 4.4K tokens were billed at the cache price

Model-family test: series and tier identification

问它自己是什么模型，看答案对不对。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Model-family identification was unstable
This test used 7.1K tokens
of which 4.5K tokens were billed at the cache price

Deep model test: family-specific boundary

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 7.1K tokens
of which 4.5K tokens were billed at the cache price

连问 5 次，重复内容有没有按缓存价计费

针对这个模型再做一项专门检测。 5 次都对。

Passed

Test details

This test ran 5 times: 5 passed.
This test used 59.1K tokens
of which 38.4K tokens were billed at the cache price

32K 上下文能力实测

32K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 32K
This test used 36.4K tokens
of which 23.6K tokens were billed at the cache price

64K 上下文能力实测

64K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 64K
This test used 66.2K tokens
of which 43.0K tokens were billed at the cache price

100K 上下文能力实测

100K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 100K
This test used 98.3K tokens
of which 63.8K tokens were billed at the cache price

200K 上下文能力实测

200K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 200K
This test used 190.0K tokens
of which 123.5K tokens were billed at the cache price

400K 上下文能力实测

400K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 400K
This test used 373.5K tokens
of which 242.7K tokens were billed at the cache price

800K 上下文能力实测

800K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 800K
This test used 740.4K tokens
of which 481.2K tokens were billed at the cache price

1M 上下文能力实测

1M 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 1M
This test used 952.3K tokens
of which 619.0K tokens were billed at the cache price

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-sonnet-5 genuine? Deep check result

Report ID:MZZR20260806025849537510Generated at：2026/08/06 10:58

Provider details AI model details

Model consistency score

Matched with issues

claude-sonnet-5 本次深度验真结果：Matched但Matched with issues（80/100）。主要问题：问它是哪个型号时答案会变；分不清自己属于哪一代模型。

Tested byMozhenzhen

Tested modelClaude Sonnet 5

Providers小火箭API

Requests42

Duration4 min

Identity verdictVerified

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.

Token cache usage test

Several runs in a row to see whether repeated content is billed at the cheaper cache rate; 0% means no cache discount this time.

Total 5 request

Share billed at the cache rate65.0%

Low cache share

Caching was used, but only a little; check your bill to see how much it really saves.

Usage this run59.1KCached tokens38.4KNon-cached tokens20.7KAverage per run11.8K

Usage reportedReportedCache billingHappenedResultNormal

Cache share by testThis is only the cache share the endpoint reported; the real discount is on the bill

Long-context test

We test how much it can read at once, then compare that with the officially stated length.

7 tiers

Verified up to 1M

All tested context tiers (32K / 64K / 100K / 200K / 400K / 800K / 1M) returned reliably.

Results for this test

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

Scoring breakdown

Matched with issues80/100

主要问题：问它是哪个型号时答案会变；分不清自己属于哪一代模型。

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 be1 / 18

问它是哪个型号时答案会变（模型代码锚定） -10分不清自己属于哪一代模型（模型族题：系列和档位识别） -7

Does it answer like the official model22 / 25

问它是哪个型号时答案会变（模型代码锚定） -3

Is usage and billing reasonable10 / 10

No deductions

Will it keep working25 / 25

No deductions

Test details

Expand any row for details; anything private has been removed

24 checks

Model list visibility

claude-sonnet-5 is visible in the API model list.

Passed

Test details

This test ran 14 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 7.2K tokens
of which 4.6K tokens were billed at the cache price

指令跟随

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 6.5K tokens
of which 4.2K tokens were billed at the cache price

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 6.5K tokens
of which 4.2K tokens were billed at the cache price

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 6.8K tokens
of which 4.4K tokens were billed at the cache price

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。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 6.6K tokens
of which 4.3K tokens were billed at the cache price

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 6.8K tokens
of which 4.3K tokens were billed at the cache price

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 6.5K tokens
of which 4.2K tokens were billed at the cache price

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 6.9K tokens
of which 4.5K tokens were billed at the cache price

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。 2 次都对。

Passed

Test details

This test ran 2 times: 2 passed.
This test used 13.1K tokens
of which 8.5K tokens were billed at the cache price

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 6.8K tokens
of which 4.3K tokens were billed at the cache price

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 6.8K tokens
of which 4.3K tokens were billed at the cache price

模型代码锚定

问它自己是什么模型，看答案对不对。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Model-code anchoring was unstable
This test used 7.0K tokens
of which 4.4K tokens were billed at the cache price

Model-family test: series and tier identification

问它自己是什么模型，看答案对不对。有几次没对上。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Model-family identification was unstable
This test used 7.1K tokens
of which 4.5K tokens were billed at the cache price

Deep model test: family-specific boundary

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 7.1K tokens
of which 4.5K tokens were billed at the cache price

连问 5 次，重复内容有没有按缓存价计费

针对这个模型再做一项专门检测。 5 次都对。

Passed

Test details

This test ran 5 times: 5 passed.
This test used 59.1K tokens
of which 38.4K tokens were billed at the cache price

32K 上下文能力实测

32K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 32K
This test used 36.4K tokens
of which 23.6K tokens were billed at the cache price

64K 上下文能力实测

64K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 64K
This test used 66.2K tokens
of which 43.0K tokens were billed at the cache price

100K 上下文能力实测

100K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 100K
This test used 98.3K tokens
of which 63.8K tokens were billed at the cache price

200K 上下文能力实测

200K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 200K
This test used 190.0K tokens
of which 123.5K tokens were billed at the cache price

400K 上下文能力实测

400K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 400K
This test used 373.5K tokens
of which 242.7K tokens were billed at the cache price

800K 上下文能力实测

800K 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 800K
This test used 740.4K tokens
of which 481.2K tokens were billed at the cache price

1M 上下文能力实测

1M 长度的内容能稳定读完，这就是本次上下文实测的依据。

Passed

Test details

This test ran 1 times: 1 passed.
Content length used: 1M
This test used 952.3K tokens
of which 619.0K tokens were billed at the cache price

Reviews

还没有人评论，来写第一条。