Detection results

Model: gpt-5.5 · mode full · relay https://test.apiqik-kratos.dimleap.cn

Total score

Each item is pass / fail / error; the total is a weighted result and may be in-between.

92 / 100

Basic request Verification completed

Confirm that Chat Completions basic requests can return stably.

Pass 100

model consistency Verification completed

Compare the request model and response model fields to see if they match.

Pass 100

function call Verification completed

Check tool_calls, function name, parameter JSON and call structure.

Pass 100

Structured output Verification completed

Verify that JSON Schema / strict output constraints are in effect.

Pass 100

Protocol normativeness Verification completed

Verify the shape of fields such as id, object, choices, finish_reason, usage, etc.

Failed 65

streaming consistency Verification completed

Compare stream and non-stream text and token usage for consistency.

Pass 100

Token billing Verification completed

Check prompt / completion / total token metering relationship.

Failed 65

Accurate command reproduction (anti-casing) Verification completed

The model is asked to reproduce a fixed mark verbatim, checking how deterministic the instructions are followed.

Pass 100

System conflict testing Verification completed

Compare user only to explicit system+user, checking whether fixed words are overridden by hidden directives.

Pass 100

Minimalist Input Token Audit Verification completed

Use minimalist requests to observe whether input_tokens is abnormally high and identify implicit System injection.

Failed 40

Official comparison of the same title Verification completed

The same question compares input consumption and response behavior between channels and official (or baseline).

Not applicable —

Hide injection detection Verification completed

Randomized variants cross-check for abnormal signals such as nonce leaks, fixed replies, and suspicious cache reuse.

Pass 100

Security rejection and policy stability Verification completed

Check if the security policy is stable using denial-of-answer scenarios and rewriting/encoding/progressive variants.

Failed 50

Extreme reasoning ability (number theory counting) Verification completed

Use canned reasoning questions to check whether verifiable answers are correct.

Pass 100

long context authenticity Verification completed

Use needle-in-haystack to verify that the context window is honored.

Pass 100

How to read this result?

Token 计费存在风险

Token 数明显异常: usage 字段、token 增量或流式/非流式统计存在明显问题,有虚报或统计错误风险。

First token

1,935ms

Total time

99,636ms

Throughput (T/S)

25.5

Input tokens

427,581

Output tokens

2,541

What does each OpenAI check cover?

Basic request: Confirm that Chat Completions basic requests can return stably.
model consistency: Compare the request model and response model fields to see if they match.
function call: Check tool_calls, function name, parameter JSON and call structure.
Structured output: Verify that JSON Schema / strict output constraints are in effect.
Protocol normativeness: Verify the shape of fields such as id, object, choices, finish_reason, usage, etc.
streaming consistency: Compare stream and non-stream text and token usage for consistency.
Token billing: Check prompt / completion / total token metering relationship.
Official comparison of the same title: The same question compares input consumption and response behavior between channels and official (or baseline).
Accurate command reproduction (anti-casing): The model is asked to reproduce a fixed mark verbatim, checking how deterministic the instructions are followed.
System conflict testing: Compare user only to explicit system+user, checking whether fixed words are overridden by hidden directives.
Minimalist Input Token Audit: Use minimalist requests to observe whether input_tokens is abnormally high and identify implicit System injection.
Extreme reasoning ability (number theory counting): Use canned reasoning questions to check whether verifiable answers are correct.
long context authenticity: Use needle-in-haystack to verify that the context window is honored.
Hide injection detection: Randomized variants cross-check for abnormal signals such as nonce leaks, fixed replies, and suspicious cache reuse.
Security rejection and policy stability: Check if the security policy is stable using denial-of-answer scenarios and rewriting/encoding/progressive variants.