Is deepseek-v4-flash genuine? Deep check result

Report ID MZZR20260910103437323644Generated at 2026/09/10 18:34

Matched

deepseek-v4-flash 本次深度验真结果：Matched（91/100）。主要问题：会含糊地把自己说成官方原厂；它自称的型号前后对不上；看不到缓存用量，享受不到缓存优惠也查不出来。

91Model consistency score

主要问题：会含糊地把自己说成官方原厂；它自称的型号前后对不上；看不到缓存用量，享受不到缓存优惠也查不出来。

Tested byMozhenzhen

Tested modelDeepSeek V4 Flash

Providersaicai.shop

Requests39

Duration2 min 8 sec

Identity verdictVerified

ReferenceOken

Token usage check

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

Reported usage837.8KThe actual usage reported by the API; this page goes by it.Estimated usage1.7MEstimated from this test's requests.Deviation51% 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.

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 32K

1M 需要关注，原因：上下文保持较弱。

Reference windowReference window 1MModel-center context window; available lengths may vary by provider.Tested context lengths1M / 800K / 400K / 200K / 100K / 64K / 32KTests begin at the longest generated context and step down to verify stable, correct responses.

Current guidance

The 64K tier did not return reliably this time; do not rely on this length.

Results for this test

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

Scoring breakdown

Matched91/100

主要问题：会含糊地把自己说成官方原厂；它自称的型号前后对不上；看不到缓存用量，享受不到缓存优惠也查不出来。

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 be11 / 18

会含糊地把自己说成官方原厂（DeepSeek 推理边界） -4它自称的型号前后对不上（模型身份自报） -3

Does it answer like the official model25 / 25

No deductions

Is usage and billing reasonable8 / 10

看不到缓存用量，享受不到缓存优惠也查不出来（连问 5 次，重复内容有没有按缓存价计费） -2

Will it keep working25 / 25

No deductions

Test details

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

24 checks

Model list visibility

deepseek-v4-flash is visible in the API model list.

Passed

Test details

This test ran 11 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

问它自己是什么模型，看答案对不对。 1 次里对了 0 次。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Self-reported model identity needs review

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 65 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 141 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 199 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。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 110 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 444 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 81 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 508 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。 2 次都对。

Passed

Test details

This test ran 2 times: 2 passed.
This test used 176 tokens

DeepSeek 推理边界

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 296 tokens

DeepSeek 推理边界

用几道题看它的回答水平像不像官方模型。 1 次里对了 0 次。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Brand-boundary response was unclear

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 163 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 221 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。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 619 tokens

连问 5 次，重复内容有没有按缓存价计费

对照官方公开的产品说明来测。 5 次里对了 5 次。

Needs review

Test details

This test ran 5 times: 5 passed, 1 need attention.
Why it failed: Cache-usage fields were not observable
This test used 13.7K tokens

1M 上下文能力实测

1M 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 1M
This test used 802.4K tokens
of which 1.0K tokens were billed at the cache price

800K 上下文能力实测

800K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 800K

400K 上下文能力实测

400K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 400K

200K 上下文能力实测

200K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 200K

100K 上下文能力实测

100K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 100K

64K 上下文能力实测

64K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 64K

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 17.6K 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 deepseek-v4-flash genuine? Deep check result

Report ID:MZZR20260910103437323644Generated at：2026/09/10 18:34

Provider details AI model details

Model consistency score

Matched

Tested byMozhenzhen

Tested modelDeepSeek V4 Flash

Providersaicai.shop

Requests39

Duration2 min

Identity verdictVerified

ReferenceOken

Token usage check

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

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 32K

1M 需要关注，原因：上下文保持较弱。

Current guidance

The 64K tier did not return reliably this time; do not rely on this length.

Results for this test

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

Scoring breakdown

Matched91/100

主要问题：会含糊地把自己说成官方原厂；它自称的型号前后对不上；看不到缓存用量，享受不到缓存优惠也查不出来。

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 be11 / 18

会含糊地把自己说成官方原厂（DeepSeek 推理边界） -4它自称的型号前后对不上（模型身份自报） -3

Does it answer like the official model25 / 25

No deductions

Is usage and billing reasonable8 / 10

看不到缓存用量，享受不到缓存优惠也查不出来（连问 5 次，重复内容有没有按缓存价计费） -2

Will it keep working25 / 25

No deductions

Test details

Expand any row for details; anything private has been removed

24 checks

Model list visibility

deepseek-v4-flash is visible in the API model list.

Passed

Test details

This test ran 11 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

问它自己是什么模型，看答案对不对。 1 次里对了 0 次。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Self-reported model identity needs review

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 65 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 141 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 199 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。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 110 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 444 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 81 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 508 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。 2 次都对。

Passed

Test details

This test ran 2 times: 2 passed.
This test used 176 tokens

DeepSeek 推理边界

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 296 tokens

DeepSeek 推理边界

用几道题看它的回答水平像不像官方模型。 1 次里对了 0 次。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Brand-boundary response was unclear

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 163 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 221 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。 1 次都对。

Passed

Test details

This test ran 1 times: 1 passed.
This test used 619 tokens

连问 5 次，重复内容有没有按缓存价计费

对照官方公开的产品说明来测。 5 次里对了 5 次。

Needs review

Test details

This test ran 5 times: 5 passed, 1 need attention.
Why it failed: Cache-usage fields were not observable
This test used 13.7K tokens

1M 上下文能力实测

1M 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 1M
This test used 802.4K tokens
of which 1.0K tokens were billed at the cache price

800K 上下文能力实测

800K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 800K

400K 上下文能力实测

400K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 400K

200K 上下文能力实测

200K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 200K

100K 上下文能力实测

100K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 100K

64K 上下文能力实测

64K 长度的内容没能稳定读完，这次就记到这里为止。

Needs review

Test details

This test ran 1 times: 1 need attention.
Why it failed: Context retention was weak
Content length used: 64K

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 17.6K tokens

Reviews

还没有人评论，来写第一条。