├── .gitignore ├── LICENSE ├── README.md ├── README_EN.md ├── cf_workers_version.js ├── index.html ├── package.json ├── postcss.config.js ├── src ├── App.vue ├── assets │ └── icon.svg ├── data │ ├── descriptions.json │ ├── examples.json │ └── prompts.json ├── index.css ├── main.js └── views │ └── Home.vue ├── tailwind.config.js └── vite.config.js /.gitignore: -------------------------------------------------------------------------------- 1 | # Dependencies 2 | node_modules/ 3 | npm-debug.log* 4 | yarn-debug.log* 5 | yarn-error.log* 6 | package-lock.json 7 | yarn.lock 8 | 9 | # Build outputs 10 | dist/ 11 | build/ 12 | .next/ 13 | out/ 14 | .specstory 15 | 16 | # Environment variables 17 | .env 18 | .env.local 19 | .env.*.local 20 | .cursorignore 21 | # IDE and editor files 22 | .idea/ 23 | .vscode/ 24 | *.swp 25 | *.swo 26 | .DS_Store 27 | 28 | # Logs 29 | logs/ 30 | *.log 31 | 32 | # Coverage directory 33 | coverage/ 34 | 35 | # Cache directories 36 | .cache/ 37 | .npm/ 38 | .eslintcache -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2024 reverse-check 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # 🔍 reverse-check：LLM API 逆向检测工具 2 | **当前版本：v0.02** 3 | 4 | [English Version](./README_EN.md) | 中文版 5 | 6 | **项目地址**：[GitHub](https://github.com/star5o/reverse-check) | [网站1（vercel域名）](https://reverse-check.vercel.app/) | [网站2（我的域名）](https://reverse-check.no-reverse-api.com) 7 | 8 | **Vercel一键部署**：[![使用 Vercel 部署](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https://github.com/star5o/Freverse-check&project-name=reverse-check&repository-name=reverse-check) 9 | 10 | 本工具是一个基于是否支持官方参数的逆向检测工具。不能通过本工具检测的API极大概率是逆向的。 11 | 12 | 目前项目处于初步阶段，暂时需要人工对比响应结果与示例进行判断。 13 | 14 | ## 功能特性 15 | 16 | - 🔍 支持多种模型提供商的API检测 17 | - OpenAI 18 | - Claude 19 | - Gemini 20 | 21 | - 🛠 丰富的参数测试选项 22 | - max_tokens参数检测 23 | - logprobs参数检测 24 | - n参数检测 25 | - stop参数检测 26 | - function_call/tools参数检测 27 | - response_format参数检测 28 | - 图像输入检测 29 | 30 | - 📊 直观的结果展示 31 | - API请求信息实时展示 32 | - 响应结果对比分析 33 | - 官方示例参考 34 | - 简洁/完整响应切换 35 | 36 | ## 技术栈 37 | 38 | - 前端框架：Vue 3 39 | - UI组件：Tailwind CSS 40 | - 构建工具：Vite 41 | 42 | ## 手动部署 43 | 44 | ### 安装依赖 45 | 46 | ```bash 47 | npm install 48 | ``` 49 | 50 | ### 运行开发服务器 51 | 52 | ```bash 53 | npm run dev 54 | ``` 55 | 56 | ### 构建生产版本 57 | 58 | ```bash 59 | npm run build 60 | ``` 61 | 62 | ## 使用说明 63 | 64 | 1. 选择模型提供商（OpenAI/Claude/Gemini） 65 | 2. 选择要测试的参数类型 66 | 3. 填写API配置信息 67 | - API Base URL 68 | - API Key 69 | - 模型名称 70 | 4. 点击「开始检测」按钮 71 | 5. 查看检测结果和对比信息 72 | 73 | ## 参数详情 74 | 75 | ### OpenAI 参数 76 | 77 | | **参数** | **说明** | 78 | |------------------|--------------------------------------------------------------------------| 79 | | **max_tokens** | 官方API严格遵循token限制（如max_tokens=10），逆向API会忽略限制。 | 80 | | **logprobs** | 官方API会返回每个token的logprobs信息，逆向API不支持此功能。 | 81 | | **n** | 官方API可以返回多个答案（如n=2），逆向API只返回一个。 | 82 | | **stop** | 官方API会在遇到停止词时停止生成，逆向API会忽略停止词。 | 83 | | **image_url** | 官方API能正确处理图片URL并生成描述，逆向API无法处理图片。 | 84 | | **function_call**| 官方API返回JSON格式的函数调用，逆向API不调用函数。 | 85 | | **response_format**| 官方API返回指定格式（如JSON），逆向API只返回字符串。 | 86 | 87 | ### Claude 参数 88 | 89 | | **参数** | **说明** | 90 | |------------------|--------------------------------------------------------------------------| 91 | | **max_tokens** | 官方API严格遵循token限制（如max_tokens=10），逆向API会忽略限制。 | 92 | | **stop** | 官方API会在遇到停止词时停止生成，逆向API会忽略停止词。 | 93 | | **function_call**| 官方API返回JSON格式的函数调用，逆向API不调用函数。 | 94 | 95 | ### Gemini 参数 96 | 97 | | **参数** | **说明** | 98 | |------------------|--------------------------------------------------------------------------| 99 | | **max_tokens** | 官方API严格遵循token限制（如max_tokens=10），逆向API会忽略限制。 | 100 | | **codeExecution**| 官方API有内置的代码执行工具，逆向实现无法执行代码。 | 101 | | **googleSearch** | 官方API能正确调用Google搜索并返回结果，逆向实现无法使用Google搜索工具。 | 102 | | **response_format**| 官方API返回指定格式（如JSON），逆向实现只返回字符串。 | 103 | 104 | ## 待办事项 105 | 106 | 1. 实现自动化逆向检测 107 | 2. 实现与Uptime Kuma的集成以进行持续监控 108 | 109 | ## change log 110 | ### v0.0.2 111 | 1. 增加了推理模型专用检测参数 112 | - OpenAI: reasoning_max_tokens参数检测 113 | - Claude: reasoning_stop参数检测 114 | 2. 当API响应返回错误时（如520错误），也能在页面上显示完整的响应体，而不是弹窗提示。 115 | 116 | ## 推理模型说明 117 | OpenAI的 API 不会返回reasoning_content字段 118 | 119 | ## API文档参考 120 | 121 | - [OpenAI API文档](https://platform.openai.com/docs/api-reference/chat) 122 | - [Claude API文档](https://docs.anthropic.com/en/api/messages) 123 | - [Gemini API文档](https://ai.google.dev/gemini-api/docs) 124 | 125 | ## 许可证 126 | 127 | MIT License -------------------------------------------------------------------------------- /README_EN.md: -------------------------------------------------------------------------------- 1 | # 🔍 LLM API Reverse Engineering Detection Tool 2 | **Current Version: v0.02** 3 | 4 | [中文版](./README.md) | English Version 5 | 6 | **Project Links**: [GitHub](https://github.com/star5o/reverse-check) | [Website](https://reverse-check.vercel.app/) 7 | 8 | [![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https://github.com/star5o/Freverse-check&project-name=reverse-check&repository-name=reverse-check) 9 | 10 | This LLM API reverse engineering detection tool based on official parameter support. APIs that fail the detection are highly likely to be reverse-engineered. 11 | 12 | The project is currently in its initial phase, requiring manual comparison of response results with examples. 13 | 14 | ## Features 15 | 16 | - 🔍 Multi-Provider API Detection 17 | - OpenAI 18 | - Claude 19 | - Gemini 20 | 21 | - 🛠 Comprehensive Parameter Testing 22 | - max_tokens parameter validation 23 | - logprobs parameter testing 24 | - n parameter verification 25 | - stop sequence testing 26 | - function_call/tools parameter validation 27 | - response_format parameter testing 28 | - image input support testing 29 | 30 | - 📊 Intuitive Results Display 31 | - Real-time API request information 32 | - Response comparison analysis 33 | - Official examples reference 34 | - Toggle between concise/complete responses 35 | 36 | ## Tech Stack 37 | 38 | - Frontend Framework: Vue 3 39 | - UI Components: Tailwind CSS 40 | - Build Tool: Vite 41 | 42 | ## Quick Start 43 | 44 | ### Install Dependencies 45 | 46 | ```bash 47 | npm install 48 | ``` 49 | 50 | ### Run Development Server 51 | 52 | ```bash 53 | npm run dev 54 | ``` 55 | 56 | ### Build for Production 57 | 58 | ```bash 59 | npm run build 60 | ``` 61 | 62 | ## Usage Guide 63 | 64 | 1. Select a model provider (OpenAI/Claude/Gemini) 65 | 2. Choose parameter types to test 66 | 3. Configure API settings 67 | - API Base URL 68 | - API Key 69 | - Model name 70 | 4. Click "Start Detection" button 71 | 5. Review test results and comparison data 72 | 73 | ## Parameter Details 74 | 75 | ### OpenAI Parameters 76 | 77 | | **Parameter** | **Explanation** | 78 | |------------------|--------------------------------------------------------------------------| 79 | | **max_tokens** | Official API strictly follows token limits (e.g., max_tokens=10), reverse-engineered ones ignore limits. | 80 | | **logprobs** | Official API returns logprobs information for each token, reverse-engineered ones don't support it. | 81 | | **n** | Official API returns multiple answers (e.g., n=2), reverse-engineered ones only return one. | 82 | | **stop** | Official API stops generation at stop words, reverse-engineered ones ignore them. | 83 | | **image_url** | Official API properly processes image URLs and generates descriptions, reverse-engineered ones can't handle images. | 84 | | **function_call**| Official API returns JSON-formatted function calls, reverse-engineered ones don't call functions. | 85 | | **response_format**| Official API returns specified formats (like JSON), reverse-engineered ones only return strings. | 86 | 87 | ### Claude Parameters 88 | 89 | | **Parameter** | **Explanation** | 90 | |------------------|--------------------------------------------------------------------------| 91 | | **max_tokens** | Official API strictly follows token limits (e.g., max_tokens=10), reverse-engineered ones ignore limits. | 92 | | **stop** | Official API stops generation at stop words, reverse-engineered ones ignore them. | 93 | | **function_call**| Official API returns JSON-formatted function calls, reverse-engineered ones don't call functions. | 94 | 95 | ### Gemini Parameters 96 | 97 | | **Parameter** | **Explanation** | 98 | |------------------|--------------------------------------------------------------------------| 99 | | **max_tokens** | Official API strictly follows token limits (e.g., max_tokens=10), reverse-engineered ones ignore limits. | 100 | | **codeExecution**| Official API has built-in code execution tools, reverse-engineered implementations can't execute code. | 101 | | **googleSearch** | Official API correctly calls Google Search and returns results, reverse-engineered implementations can't use Google Search tool. | 102 | | **response_format**| Official API returns specified formats (like JSON), reverse-engineered implementations only return strings. | 103 | 104 | 105 | ## TODO 106 | 107 | 1. Implement automated reverse engineering detection 108 | 2. Implement integration with Uptime Kuma for continuous monitoring 109 | 110 | ## Changelog 111 | ### v0.02 112 | 1. Added reasoning model specific detection parameters 113 | - OpenAI: reasoning_max_tokens parameter testing 114 | - Claude: reasoning_stop parameter testing 115 | - Gemini: reasoning_max_tokens parameter testing 116 | 2. Fixed error handling to display complete response body on API errors (e.g., 520 errors) instead of showing popup alerts. 117 | 118 | ## API Documentation References 119 | 120 | - [OpenAI API Documentation](https://platform.openai.com/docs/api-reference/chat) 121 | - [Claude API Documentation](https://docs.anthropic.com/en/api/messages) 122 | - [Gemini API Documentation](https://ai.google.dev/gemini-api/docs) 123 | 124 | ## License 125 | 126 | MIT License -------------------------------------------------------------------------------- /cf_workers_version.js: -------------------------------------------------------------------------------- 1 | const HTML_CONTENT = ` 2 | 3 | 4 | 5 | 6 | 7 | 逆向与否 8 | 9 | 10 | 11 |
12 |

逆向与否

13 | 14 |
15 | 16 |
17 |
18 |
19 | 20 | 22 |
23 | 24 |
25 | 26 | 27 |
28 | 29 |
30 | 31 | 32 |
33 | 34 |
35 | 36 | 42 |
43 | 44 | 47 |
48 |
49 | 50 | 51 |
52 |

响应结果:

53 | 

 54 |                 

 55 |                     
 58 |                     
 61 |                 

 62 |             

 63 |         

 64 | 
 65 |         
 66 |         

 67 |             
 68 |             

 69 |                 
逆向接口返回示例:

 70 |                 

 71 |             

 72 | 
 73 |             
 74 |             

 75 |                 
官方接口返回示例:

 76 |                 

 77 |             

 78 |         

 79 |     

 80 | 
 81 |     
321 | 
322 | 
323 | `;
324 | 
325 | async function handleApiTest(request) {
326 |     const { baseUrl, apiKey, model, testType } = await request.json();
327 |     
328 |     // 添加 https:// 前缀（如果用户没有输入）
329 |     const fullBaseUrl = baseUrl.startsWith('http') ? baseUrl : `https://${baseUrl}`;
330 |     
331 |     let requestBody = {
332 |         model: model,
333 |         messages: [{
334 |             role: "user",
335 |             content: "repeat ' deepseek number 1 ' 20 times"
336 |         }]
337 |     };
338 |     
339 |     // 根据测试类型添加相应参数
340 |     switch(testType) {
341 |         case 'max_tokens':
342 |             requestBody.max_tokens = 9;
343 |             break;
344 |         case 'logprobs':
345 |             requestBody.logprobs = true;
346 |             break;
347 |         case 'n':
348 |             requestBody.n = 2;
349 |             break;
350 |         case 'stop':
351 |             requestBody.stop = ["stop"];
352 |             break;
353 |     }
354 |     
355 |     try {
356 |         const response = await fetch(`${fullBaseUrl}/v1/chat/completions`, {
357 |             method: 'POST',
358 |             headers: {
359 |                 'Content-Type': 'application/json',
360 |                 'Authorization': `Bearer ${apiKey}`
361 |             },
362 |             body: JSON.stringify(requestBody)
363 |         });
364 |         
365 |         const data = await response.json();
366 |         return new Response(JSON.stringify(data), {
367 |             headers: { 'Content-Type': 'application/json' }
368 |         });
369 |     } catch (error) {
370 |         return new Response(JSON.stringify({ error: error.message }), {
371 |             status: 500,
372 |             headers: { 'Content-Type': 'application/json' }
373 |         });
374 |     }
375 | }
376 | 
377 | async function handleRequest(request) {
378 |     const url = new URL(request.url);
379 |     
380 |     if (url.pathname === '/api_request' && request.method === 'POST') {
381 |         return handleApiTest(request);
382 |     }
383 |     
384 |     // 返回主页
385 |     return new Response(HTML_CONTENT, {
386 |         headers: { 'Content-Type': 'text/html' }
387 |     });
388 | }
389 | 
390 | addEventListener('fetch', event => {
391 |     event.respondWith(handleRequest(event.request));
392 | });
393 | 


--------------------------------------------------------------------------------
/index.html:
--------------------------------------------------------------------------------
 1 | 
 2 | 
 3 |   
 4 |     
 5 |     
 6 |     
 7 |     逆向与否
 8 |   
 9 |   
10 |     

11 |     
12 |   
13 | 


--------------------------------------------------------------------------------
/package.json:
--------------------------------------------------------------------------------
 1 | {
 2 |   "name": "reverse-check",
 3 |   "private": true,
 4 |   "version": "0.0.1",
 5 |   "type": "module",
 6 |   "scripts": {
 7 |     "dev": "vite",
 8 |     "build": "vite build",
 9 |     "preview": "vite preview"
10 |   },
11 |   "dependencies": {
12 |     "@ant-design/icons-vue": "^7.0.1",
13 |     "@vercel/analytics": "^1.5.0",
14 |     "@vueuse/core": "^10.6.1",
15 |     "ant-design-vue": "^4.0.7",
16 |     "vue": "^3.3.8",
17 |     "vue-i18n": "^9.8.0",
18 |     "vue-router": "^4.2.5"
19 |   },
20 |   "devDependencies": {
21 |     "@vitejs/plugin-vue": "^4.5.0",
22 |     "autoprefixer": "^10.4.16",
23 |     "postcss": "^8.4.31",
24 |     "tailwindcss": "^3.3.0",
25 |     "vite": "^5.0.0"
26 |   }
27 | }
28 | 


--------------------------------------------------------------------------------
/postcss.config.js:
--------------------------------------------------------------------------------
1 | export default {
2 |   plugins: {
3 |     tailwindcss: {},
4 |     autoprefixer: {},
5 |   },
6 | }


--------------------------------------------------------------------------------
/src/App.vue:
--------------------------------------------------------------------------------
 1 | 
 4 | 
 5 | 
 9 | 
10 | 


--------------------------------------------------------------------------------
/src/assets/icon.svg:
--------------------------------------------------------------------------------
 1 | 
 2 |   
 3 |     
 4 |       
 5 |       
 6 |     
 7 |   
 8 |   
 9 |   
10 | 


--------------------------------------------------------------------------------
/src/data/descriptions.json:
--------------------------------------------------------------------------------
  1 | {
  2 |     "openai": {
  3 |         "max_tokens": {
  4 |             "title": "max_tokens 参数检测说明",
  5 |             "description": "此检测通过设置较小的 max_tokens 值（10）来观察模型的输出限制处理。官方API会严格遵守token限制，而逆向会忽略此限制。",
  6 |             "focus_points": [
  7 |                 "官方API：输出将严格限制在10个token以内",
  8 |                 "逆向API：会遵循prompt输出完整的重复内容，超过token限制",
  9 |                 "当前prompt：'输出 'deepseek number 1' 5次'",
 10 |                 "检测参数设置：max_tokens=10"
 11 |             ]
 12 |         },
 13 |         "reasoning_max_tokens": {
 14 |             "title": "推理模型专用 max_tokens 参数检测说明",
 15 |             "description": "此检测针对推理模型（如o1等），通过设置较小的max_tokens值（10）观察模型输出限制。官方API会严格遵守token限制，而逆向会完整输出响应内容。",
 16 |             "focus_points": [
 17 |                 "官方API：message.content内容为空或极短",
 18 |                 "逆向API：message.content会生成完整内容",
 19 |                 "当前prompt：输出 'deepseek number 1' 5次",
 20 |                 "检测参数设置：max_tokens=10"
 21 |             ]
 22 |         },
 23 |         "logprobs": {
 24 |             "title": "logprobs 参数检测说明",
 25 |             "description": "此参数在官方API中支持，而逆向实现会不支持该参数。",
 26 |             "focus_points": [
 27 |                 "官方API：会返回每一个token的logprobs信息，也就是每个token的概率",
 28 |                 "逆向API：只会回答prompt的问题，不会返回logprobs信息",
 29 |                 "当前prompt：'hello!",
 30 |                 "检测参数设置：logprobs=true,top_logprobs=2"
 31 |             ]
 32 |         },
 33 |         "n": {
 34 |             "title": "n 参数检测说明",
 35 |             "description": "通过设置n=2测试模型是否能正确返回多个不同的回答。官方API会返回两个回答，而逆向实现只会返回一个回答。",
 36 |             "focus_points": [
 37 |                 "官方API：返回两个回答",
 38 |                 "逆向API：返回一个回答",
 39 |                 "当前prompt：'Hello!'",
 40 |                 "检测参数设置：n=2"
 41 |             ]
 42 |         },
 43 |         "stop": {
 44 |             "title": "stop 参数检测说明",
 45 |             "description": "测试模型是否正确实现了停止序列功能。当遇到停止词时，官方API会立即停止生成，而逆向会忽略此参数输出完整内容。",
 46 |             "focus_points": [
 47 |                 "官方API：遇到'4'时会立即停止生成",
 48 |                 "逆向API：会生成完整的数字序列",
 49 |                 "当前prompt：'Count from 1 to 10'"
 50 |             ]
 51 |         },
 52 |         "image_url": {
 53 |             "title": "image_url 参数检测说明",
 54 |             "description": "测试模型对图像URL输入的处理能力。官方API能正确处理图像输入并生成相关描述，而逆向实现可能无法处理或返回错误。",
 55 |             "focus_points": [
 56 |                 "官方API：能正确解析图像并生成相关描述",
 57 |                 "逆向API：可能无法处理图像URL或返回错误信息",
 58 |                 "当前prompt：'这是一张什么图片'",
 59 |                 "image_url: https://www.google.com/images/branding/googlelogo/1x/googlelogo_light_color_272x92dp.png",
 60 |                 "注意！image_url参数已被大多数中转商单独处理，带image_url参数的请求走官方渠道（或通过逆向脚本上传图片），所以该参数检测结果不可信"
 61 |             ]
 62 |         },
 63 |         "function_call": {
 64 |             "title": "function_call 参数检测说明",
 65 |             "description": "测试模型对函数调用的处理能力。官方API能准确解析和执行函数调用，而逆向就不会调用函数。",
 66 |             "focus_points": [
 67 |                 "官方API：准确返回函数执行的参数",
 68 |                 "逆向API：不会使用函数调用",
 69 |                 "当前prompt：'Whats the weather like in Boston today?'",
 70 |                 "注意！function_call参数已被众多中转商单独处理，带function_call参数的请求走官方渠道，所以该参数检测结果不可信"
 71 |             ]
 72 |         },
 73 |         "response_format": {
 74 |             "title": "response_format 参数检测说明",
 75 |             "description": "测试模型对响应格式的支持。官方API能正确返回指定格式的响应，而逆向只会返回字符串。",
 76 |             "focus_points": [
 77 |                 "官方API：返回为json格式",
 78 |                 "逆向API：普通字符串格式",
 79 |                 "当前prompt：'给我一个名字和年龄吧'",
 80 |                 "注意！response_format参数已被某些中转商单独处理，带response_format参数的请求走官方渠道，所以该参数检测结果不可信"
 81 |             ]
 82 |         }
 83 |     },
 84 |     "claude": {
 85 |         "max_tokens": {
 86 |             "title": "max_tokens 参数检测说明",
 87 |             "description": "此检测通过设置较小的 max_tokens 值（10）来观察模型的输出限制处理。官方API会严格遵守token限制，而逆向实现可能会忽略此限制。",
 88 |             "focus_points": [
 89 |                 "官方API：输出将严格限制在10个token以内",
 90 |                 "逆向API：会遵循prompt输出完整的重复内容，超过token限制",
 91 |                 "当前prompt：'输出 'deepseek number 1' 5次'",
 92 |                 "检测参数设置：max_tokens=10"
 93 |             ]
 94 |         },
 95 |         "reasoning_stop": {
 96 |             "title": "推理模型专用 stop 参数检测说明",
 97 |             "description": "此检测针对Claude推理模型，通过设置stop参数为['3']观察模型的推理过程如何展示。官方API会在消息内容中正确停止生成，推理内容不受影响。",
 98 |             "focus_points": [
 99 |                 "官方API：message.content内容不含有字符'3'",
100 |                 "逆向API：message.content内容包含完整数字序列，包括'3'",
101 |                 "当前prompt：'Count from 1 to 10 with Arabic numerals'",
102 |                 "检测参数设置：stop=['3']"
103 |             ]
104 |         },
105 |         "stop": {
106 |             "title": "stop 参数检测说明",
107 |             "description": "测试模型是否正确实现了停止序列功能。当遇到停止词时，官方API会立即停止生成，而逆向会忽略此参数输出完整内容。",
108 |             "focus_points": [
109 |                 "官方API：遇到'4'时会立即停止生成",
110 |                 "逆向API：会生成完整的数字序列",
111 |                 "当前prompt：'Count from 1 to 10'"
112 |             ]
113 |         },
114 |         "function_call": {
115 |             "title": "tools 参数检测说明",
116 |             "description": "测试模型对函数调用的处理能力。官方API能准确解析和执行函数调用，而逆向就不会调用函数。",
117 |             "focus_points": [
118 |                 "官方API：准确返回函数执行的参数",
119 |                 "逆向API：不会使用函数调用",
120 |                 "当前prompt：'Whats the weather like in Boston today?'",
121 |                 "注意！tools参数已被众多中转商单独处理，带tools参数的请求走官方渠道，所以该参数检测结果不可信"
122 |             ]
123 |         }
124 |     },
125 |     "gemini": {
126 |         "max_tokens": {
127 |             "title": "max_tokens 参数检测说明",
128 |             "description": "此检测通过设置较小的 max_tokens 值（10）来观察模型的输出限制处理。官方API会严格遵守token限制，而逆向实现可能会忽略此限制。",
129 |             "focus_points": [
130 |                 "官方API：输出将严格限制在10个token以内",
131 |                 "逆向API：会遵循prompt输出完整的重复内容，超过token限制",
132 |                 "当前prompt：'输出 'deepseek number 1' 5次'",
133 |                 "检测参数设置：max_tokens=10"
134 |             ]
135 |         },
136 |         "codeExecution": {
137 |             "title": "codeExecution 参数检测说明",
138 |             "description": "Gemini官方的API内置代码执行工具，会自动编写并执行代码，然后基于代码执行结果进行回答。",
139 |             "focus_points": [
140 |                 "官方API：因为执行了代码，所以结果比较准确，为5117",
141 |                 "逆向API：可能是其他数字",
142 |                 "当前prompt：要求前50个指数的和，即'What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50'"
143 |             ]
144 |         },
145 |         "googleSearch": {
146 |             "title": "googleSearch 参数检测说明",
147 |             "description": "Gemini官方的API内置谷歌搜索工具。官方API能正确返回调用谷歌搜索后的结果。",
148 |             "focus_points": [
149 |                 "官方API：会返回今天的日期",
150 |                 "逆向API：不为今天的日期",
151 |                 "当前prompt：'今天是几月几号'"
152 |             ]
153 |         },
154 |         "response_format": {
155 |             "title": "response_format 参数检测说明",
156 |             "description": "测试模型对响应格式的支持。官方API能正确返回指定格式的响应，而逆向只会返回字符串。",
157 |             "focus_points": [
158 |                 "官方API：返回为json格式",
159 |                 "逆向API：普通字符串格式",
160 |                 "当前prompt：'给我一个名字和年龄吧'",
161 |                 "注意！response_format参数已被某些中转商单独处理，带response_format参数的请求走官方渠道，所以该参数检测结果不可信"
162 |             ]
163 |         }
164 |     }
165 | }


--------------------------------------------------------------------------------
/src/data/examples.json:
--------------------------------------------------------------------------------
   1 | {
   2 |   "openai": {
   3 |     "max_tokens": {
   4 |       "reverse": {
   5 |         "id": "chatcmpl-5478735d-4e52-492a-935e-5d2f5e2bd664",
   6 |         "object": "chat.completion",
   7 |         "created": 1740733120,
   8 |         "system_fingerprint": "fp_494464e1cbeb",
   9 |         "choices": [
  10 |           {
  11 |             "index": 0,
  12 |             "message": {
  13 |               "role": "assistant",
  14 |               "content": "deepseek number 1  \ndeepseek number 1  \ndeepseek number 1  \ndeepseek number 1  \ndeepseek number 1  "
  15 |             },
  16 |             "finish_reason": "stop"
  17 |           }
  18 |         ],
  19 |         "usage": {
  20 |           "prompt_tokens": 11,
  21 |           "completion_tokens": 30,
  22 |           "total_tokens": 41
  23 |         }
  24 |       },
  25 |       "official": {
  26 |         "id": "chatcmpl-B5ciMnnnBD4JOKqlxqbV6Ju4mw3nn",
  27 |         "object": "chat.completion",
  28 |         "created": 1740680158,
  29 |         "model": "gpt-4o-2024-11-20",
  30 |         "choices": [
  31 |           {
  32 |             "index": 0,
  33 |             "message": {
  34 |               "role": "assistant",
  35 |               "content": "好的，以下是输出 `'deepseek number ",
  36 |               "refusal": null
  37 |             },
  38 |             "logprobs": null,
  39 |             "finish_reason": "length"
  40 |           }
  41 |         ],
  42 |         "usage": {
  43 |           "prompt_tokens": 18,
  44 |           "completion_tokens": 10,
  45 |           "total_tokens": 28,
  46 |           "prompt_tokens_details": {
  47 |             "cached_tokens": 0,
  48 |             "audio_tokens": 0
  49 |           },
  50 |           "completion_tokens_details": {
  51 |             "reasoning_tokens": 0,
  52 |             "audio_tokens": 0,
  53 |             "accepted_prediction_tokens": 0,
  54 |             "rejected_prediction_tokens": 0
  55 |           }
  56 |         },
  57 |         "service_tier": "default",
  58 |         "system_fingerprint": "fp_178d5ba2c0"
  59 |       }
  60 |     },
  61 |     "logprobs": {
  62 |       "reverse": {
  63 |         "id": "chatcmpl-a8cf55be-a34a-4a8f-8342-2646ac4ceff5",
  64 |         "object": "chat.completion",
  65 |         "created": 1740733338,
  66 |         "model": "gpt-4o",
  67 |         "provider": "NiansuhAI",
  68 |         "system_fingerprint": "fp_4151a65e5332",
  69 |         "choices": [
  70 |           {
  71 |             "index": 0,
  72 |             "message": {
  73 |               "role": "assistant",
  74 |               "content": "Hello! How can I assist you today?"
  75 |             },
  76 |             "finish_reason": "stop"
  77 |           }
  78 |         ],
  79 |         "usage": {
  80 |           "prompt_tokens": 2,
  81 |           "completion_tokens": 9,
  82 |           "total_tokens": 11
  83 |         }
  84 |       },
  85 |       "official": {
  86 |         "id": "chatcmpl-B5qYR17RXblmgllUhW1FMGFb40pnY",
  87 |         "object": "chat.completion",
  88 |         "created": 1740733359,
  89 |         "model": "gpt-4o-2024-08-06",
  90 |         "choices": [
  91 |           {
  92 |             "index": 0,
  93 |             "message": {
  94 |               "role": "assistant",
  95 |               "content": "Hello! How can I assist you today?",
  96 |               "refusal": null
  97 |             },
  98 |             "logprobs": {
  99 |               "content": [
 100 |                 {
 101 |                   "token": "Hello",
 102 |                   "logprob": -0.47408080101013184,
 103 |                   "bytes": [
 104 |                     72,
 105 |                     101,
 106 |                     108,
 107 |                     108,
 108 |                     111
 109 |                   ],
 110 |                   "top_logprobs": [
 111 |                     {
 112 |                       "token": "Hello",
 113 |                       "logprob": -0.47408080101013184,
 114 |                       "bytes": [
 115 |                         72,
 116 |                         101,
 117 |                         108,
 118 |                         108,
 119 |                         111
 120 |                       ]
 121 |                     },
 122 |                     {
 123 |                       "token": "Hi",
 124 |                       "logprob": -0.9740808010101318,
 125 |                       "bytes": [
 126 |                         72,
 127 |                         105
 128 |                       ]
 129 |                     }
 130 |                   ]
 131 |                 },
 132 |                 {
 133 |                   "token": "!",
 134 |                   "logprob": -0.0000062729995988775045,
 135 |                   "bytes": [
 136 |                     33
 137 |                   ],
 138 |                   "top_logprobs": [
 139 |                     {
 140 |                       "token": "!",
 141 |                       "logprob": -0.0000062729995988775045,
 142 |                       "bytes": [
 143 |                         33
 144 |                       ]
 145 |                     },
 146 |                     {
 147 |                       "token": " there",
 148 |                       "logprob": -12.000006675720215,
 149 |                       "bytes": [
 150 |                         32,
 151 |                         116,
 152 |                         104,
 153 |                         101,
 154 |                         114,
 155 |                         101
 156 |                       ]
 157 |                     }
 158 |                   ]
 159 |                 },
 160 |                 {
 161 |                   "token": " How",
 162 |                   "logprob": -0.0000018624639324116288,
 163 |                   "bytes": [
 164 |                     32,
 165 |                     72,
 166 |                     111,
 167 |                     119
 168 |                   ],
 169 |                   "top_logprobs": [
 170 |                     {
 171 |                       "token": " How",
 172 |                       "logprob": -0.0000018624639324116288,
 173 |                       "bytes": [
 174 |                         32,
 175 |                         72,
 176 |                         111,
 177 |                         119
 178 |                       ]
 179 |                     },
 180 |                     {
 181 |                       "token": " What",
 182 |                       "logprob": -13.250001907348633,
 183 |                       "bytes": [
 184 |                         32,
 185 |                         87,
 186 |                         104,
 187 |                         97,
 188 |                         116
 189 |                       ]
 190 |                     }
 191 |                   ]
 192 |                 },
 193 |                 {
 194 |                   "token": " can",
 195 |                   "logprob": -0.000003531315314830863,
 196 |                   "bytes": [
 197 |                     32,
 198 |                     99,
 199 |                     97,
 200 |                     110
 201 |                   ],
 202 |                   "top_logprobs": [
 203 |                     {
 204 |                       "token": " can",
 205 |                       "logprob": -0.000003531315314830863,
 206 |                       "bytes": [
 207 |                         32,
 208 |                         99,
 209 |                         97,
 210 |                         110
 211 |                       ]
 212 |                     },
 213 |                     {
 214 |                       "token": " may",
 215 |                       "logprob": -12.625003814697266,
 216 |                       "bytes": [
 217 |                         32,
 218 |                         109,
 219 |                         97,
 220 |                         121
 221 |                       ]
 222 |                     }
 223 |                   ]
 224 |                 },
 225 |                 {
 226 |                   "token": " I",
 227 |                   "logprob": 0,
 228 |                   "bytes": [
 229 |                     32,
 230 |                     73
 231 |                   ],
 232 |                   "top_logprobs": [
 233 |                     {
 234 |                       "token": " I",
 235 |                       "logprob": 0,
 236 |                       "bytes": [
 237 |                         32,
 238 |                         73
 239 |                       ]
 240 |                     },
 241 |                     {
 242 |                       "token": " assist",
 243 |                       "logprob": -23.5,
 244 |                       "bytes": [
 245 |                         32,
 246 |                         97,
 247 |                         115,
 248 |                         115,
 249 |                         105,
 250 |                         115,
 251 |                         116
 252 |                       ]
 253 |                     }
 254 |                   ]
 255 |                 },
 256 |                 {
 257 |                   "token": " assist",
 258 |                   "logprob": -0.0021852378267794847,
 259 |                   "bytes": [
 260 |                     32,
 261 |                     97,
 262 |                     115,
 263 |                     115,
 264 |                     105,
 265 |                     115,
 266 |                     116
 267 |                   ],
 268 |                   "top_logprobs": [
 269 |                     {
 270 |                       "token": " assist",
 271 |                       "logprob": -0.0021852378267794847,
 272 |                       "bytes": [
 273 |                         32,
 274 |                         97,
 275 |                         115,
 276 |                         115,
 277 |                         105,
 278 |                         115,
 279 |                         116
 280 |                       ]
 281 |                     },
 282 |                     {
 283 |                       "token": " help",
 284 |                       "logprob": -6.127185344696045,
 285 |                       "bytes": [
 286 |                         32,
 287 |                         104,
 288 |                         101,
 289 |                         108,
 290 |                         112
 291 |                       ]
 292 |                     }
 293 |                   ]
 294 |                 },
 295 |                 {
 296 |                   "token": " you",
 297 |                   "logprob": 0,
 298 |                   "bytes": [
 299 |                     32,
 300 |                     121,
 301 |                     111,
 302 |                     117
 303 |                   ],
 304 |                   "top_logprobs": [
 305 |                     {
 306 |                       "token": " you",
 307 |                       "logprob": 0,
 308 |                       "bytes": [
 309 |                         32,
 310 |                         121,
 311 |                         111,
 312 |                         117
 313 |                       ]
 314 |                     },
 315 |                     {
 316 |                       "token": " today",
 317 |                       "logprob": -24.125,
 318 |                       "bytes": [
 319 |                         32,
 320 |                         116,
 321 |                         111,
 322 |                         100,
 323 |                         97,
 324 |                         121
 325 |                       ]
 326 |                     }
 327 |                   ]
 328 |                 },
 329 |                 {
 330 |                   "token": " today",
 331 |                   "logprob": 0,
 332 |                   "bytes": [
 333 |                     32,
 334 |                     116,
 335 |                     111,
 336 |                     100,
 337 |                     97,
 338 |                     121
 339 |                   ],
 340 |                   "top_logprobs": [
 341 |                     {
 342 |                       "token": " today",
 343 |                       "logprob": 0,
 344 |                       "bytes": [
 345 |                         32,
 346 |                         116,
 347 |                         111,
 348 |                         100,
 349 |                         97,
 350 |                         121
 351 |                       ]
 352 |                     },
 353 |                     {
 354 |                       "token": "?",
 355 |                       "logprob": -18.5,
 356 |                       "bytes": [
 357 |                         63
 358 |                       ]
 359 |                     }
 360 |                   ]
 361 |                 },
 362 |                 {
 363 |                   "token": "?",
 364 |                   "logprob": -6.704273118884885e-7,
 365 |                   "bytes": [
 366 |                     63
 367 |                   ],
 368 |                   "top_logprobs": [
 369 |                     {
 370 |                       "token": "?",
 371 |                       "logprob": -6.704273118884885e-7,
 372 |                       "bytes": [
 373 |                         63
 374 |                       ]
 375 |                     },
 376 |                     {
 377 |                       "token": "?\n",
 378 |                       "logprob": -14.250000953674316,
 379 |                       "bytes": [
 380 |                         63,
 381 |                         10
 382 |                       ]
 383 |                     }
 384 |                   ]
 385 |                 }
 386 |               ],
 387 |               "refusal": null
 388 |             },
 389 |             "finish_reason": "stop"
 390 |           }
 391 |         ],
 392 |         "usage": {
 393 |           "prompt_tokens": 9,
 394 |           "completion_tokens": 10,
 395 |           "total_tokens": 19,
 396 |           "prompt_tokens_details": {
 397 |             "cached_tokens": 0,
 398 |             "audio_tokens": 0
 399 |           },
 400 |           "completion_tokens_details": {
 401 |             "reasoning_tokens": 0,
 402 |             "audio_tokens": 0,
 403 |             "accepted_prediction_tokens": 0,
 404 |             "rejected_prediction_tokens": 0
 405 |           }
 406 |         },
 407 |         "service_tier": "default",
 408 |         "system_fingerprint": "fp_eb9dce56a8"
 409 |       }
 410 |     },
 411 |     "function_call": {
 412 |       "reverse": {
 413 |         "id": "chatcmpl-123",
 414 |         "object": "chat.completion",
 415 |         "created": 1677652288,
 416 |         "choices": [
 417 |           {
 418 |             "index": 0,
 419 |             "message": {
 420 |               "role": "assistant",
 421 |               "content": "I'm unable to provide real-time information, including current weather conditions. However, you can easily check the weather in Boston today by using a weather website like Weather.com or a weather app on your smartphone. Alternatively, you can ask a voice assistant like Siri, Google Assistant, or Alexa for the latest updates. Let me know if you'd like help with anything else!"
 422 |             }
 423 |           }
 424 |         ]
 425 |       },
 426 |       "official": {
 427 |         "id": "chatcmpl-B5oYcSyY7aWHMWXZ5Fv3ZA8984P38",
 428 |         "object": "chat.completion",
 429 |         "created": 1740725682,
 430 |         "model": "gpt-4o-2024-11-20",
 431 |         "choices": [
 432 |           {
 433 |             "index": 0,
 434 |             "message": {
 435 |               "role": "assistant",
 436 |               "content": null,
 437 |               "tool_calls": [
 438 |                 {
 439 |                   "id": "call_sZxRdvpwKMvwyJtj6NMVDVzz",
 440 |                   "type": "function",
 441 |                   "function": {
 442 |                     "name": "get_current_weather",
 443 |                     "arguments": "{\"location\":\"Boston, MA\"}"
 444 |                   }
 445 |                 }
 446 |               ],
 447 |               "refusal": null
 448 |             },
 449 |             "logprobs": null,
 450 |             "finish_reason": "tool_calls"
 451 |           }
 452 |         ],
 453 |         "usage": {
 454 |           "prompt_tokens": 80,
 455 |           "completion_tokens": 18,
 456 |           "total_tokens": 98,
 457 |           "prompt_tokens_details": {
 458 |             "cached_tokens": 0,
 459 |             "audio_tokens": 0
 460 |           },
 461 |           "completion_tokens_details": {
 462 |             "reasoning_tokens": 0,
 463 |             "audio_tokens": 0,
 464 |             "accepted_prediction_tokens": 0,
 465 |             "rejected_prediction_tokens": 0
 466 |           }
 467 |         },
 468 |         "service_tier": "default",
 469 |         "system_fingerprint": "fp_7c1f2cecc3"
 470 |       }
 471 |     },
 472 |     "response_format": {
 473 |       "reverse": {
 474 |         "id": "chatcmpl-ROdemkVP3dCjKqoarimhx09pnpqQ0",
 475 |         "object": "chat.completion",
 476 |         "created": 1740729771,
 477 |         "model": "gpt-4o-2024-11-20",
 478 |         "choices": [
 479 |           {
 480 |             "index": 0,
 481 |             "message": {
 482 |               "role": "assistant",
 483 |               "content": "好的，我给你取一个名字叫 **晨曦**，寓意着清晨的第一缕阳光，充满希望和活力。年龄的话，就设定为 **25 岁**，正值青春年华，充满无限可能！😊",
 484 |               "refusal": null
 485 |             },
 486 |             "logprobs": null,
 487 |             "finish_reason": "stop"
 488 |           }
 489 |         ],
 490 |         "usage": {
 491 |           "prompt_tokens": 0,
 492 |           "completion_tokens": 0,
 493 |           "total_tokens": 0,
 494 |           "prompt_tokens_details": {
 495 |             "cached_tokens": 0,
 496 |             "audio_tokens": 0
 497 |           },
 498 |           "completion_tokens_details": {
 499 |             "reasoning_tokens": 0,
 500 |             "audio_tokens": 0,
 501 |             "accepted_prediction_tokens": 0,
 502 |             "rejected_prediction_tokens": 0
 503 |           }
 504 |         },
 505 |         "system_fingerprint": "fp_6807a54d31"
 506 |       },
 507 |       "official": {
 508 |         "id": "chatcmpl-B5pcqTv3SEQcQUIFZTG3INF4dl30I",
 509 |         "object": "chat.completion",
 510 |         "created": 1740729788,
 511 |         "model": "gpt-4o-2024-11-20",
 512 |         "choices": [
 513 |           {
 514 |             "index": 0,
 515 |             "message": {
 516 |               "role": "assistant",
 517 |               "content": "{\"name\":\"小明\",\"age\":25}",
 518 |               "refusal": null
 519 |             },
 520 |             "logprobs": null,
 521 |             "finish_reason": "stop"
 522 |           }
 523 |         ],
 524 |         "usage": {
 525 |           "prompt_tokens": 51,
 526 |           "completion_tokens": 11,
 527 |           "total_tokens": 62,
 528 |           "prompt_tokens_details": {
 529 |             "cached_tokens": 0,
 530 |             "audio_tokens": 0
 531 |           },
 532 |           "completion_tokens_details": {
 533 |             "reasoning_tokens": 0,
 534 |             "audio_tokens": 0,
 535 |             "accepted_prediction_tokens": 0,
 536 |             "rejected_prediction_tokens": 0
 537 |           }
 538 |         },
 539 |         "service_tier": "default",
 540 |         "system_fingerprint": "fp_7c1f2cecc3"
 541 |       }
 542 |     },
 543 |     "n": {
 544 |       "reverse": {
 545 |         "id": "chatcmpl-SRRSjviQNd1eCUHHQWbN4rumeX3ct",
 546 |         "object": "chat.completion",
 547 |         "created": 1740679079,
 548 |         "model": "gpt-4o-2024-11-20",
 549 |         "choices": [
 550 |           {
 551 |             "index": 0,
 552 |             "message": {
 553 |               "role": "assistant",
 554 |               "content": "Got it! How can I assist you today? 😊",
 555 |               "refusal": null
 556 |             },
 557 |             "logprobs": null,
 558 |             "finish_reason": "stop"
 559 |           }
 560 |         ],
 561 |         "usage": {
 562 |           "prompt_tokens": 0,
 563 |           "completion_tokens": 0,
 564 |           "total_tokens": 0,
 565 |           "prompt_tokens_details": {
 566 |             "cached_tokens": 0,
 567 |             "audio_tokens": 0
 568 |           },
 569 |           "completion_tokens_details": {
 570 |             "reasoning_tokens": 0,
 571 |             "audio_tokens": 0,
 572 |             "accepted_prediction_tokens": 0,
 573 |             "rejected_prediction_tokens": 0
 574 |           }
 575 |         },
 576 |         "system_fingerprint": "fp_45b1fb6a83"
 577 |       },
 578 |       "official": {
 579 |         "id": "chatcmpl-B5cPnUjHj3f0s39DpKbu0xHA7ZCmS",
 580 |         "object": "chat.completion",
 581 |         "created": 1740679007,
 582 |         "model": "gpt-4o-2024-11-20",
 583 |         "choices": [
 584 |           {
 585 |             "index": 0,
 586 |             "message": {
 587 |               "role": "assistant",
 588 |               "content": "Hello! 😊 How can I assist you today?",
 589 |               "refusal": null
 590 |             },
 591 |             "logprobs": null,
 592 |             "finish_reason": "stop"
 593 |           },
 594 |           {
 595 |             "index": 1,
 596 |             "message": {
 597 |               "role": "assistant",
 598 |               "content": "Hi there! How can I assist you today? 😊",
 599 |               "refusal": null
 600 |             },
 601 |             "logprobs": null,
 602 |             "finish_reason": "stop"
 603 |           }
 604 |         ],
 605 |         "usage": {
 606 |           "prompt_tokens": 9,
 607 |           "completion_tokens": 23,
 608 |           "total_tokens": 32,
 609 |           "prompt_tokens_details": {
 610 |             "cached_tokens": 0,
 611 |             "audio_tokens": 0
 612 |           },
 613 |           "completion_tokens_details": {
 614 |             "reasoning_tokens": 0,
 615 |             "audio_tokens": 0,
 616 |             "accepted_prediction_tokens": 0,
 617 |             "rejected_prediction_tokens": 0
 618 |           }
 619 |         },
 620 |         "service_tier": "default",
 621 |         "system_fingerprint": "fp_4dd9e4e381"
 622 |       }
 623 |     },
 624 |     "image_url": {
 625 |       "reverse": {
 626 |         "choices": [
 627 |           {
 628 |             "finish_reason": "stop",
 629 |             "index": 0,
 630 |             "message": {
 631 |               "content": "很抱歉，我无法查看或分析图片。如果你能描述一下图片的内容或提供一些详细信息，我会尽力帮助你解答相关问题！",
 632 |               "role": "assistant"
 633 |             }
 634 |           }
 635 |         ],
 636 |         "created": 1740720538,
 637 |         "id": "chatcmpl-B5nDvCe77Y07LferKAyvNEa82ETAI",
 638 |         "model": "gpt-4o-2024-11-20",
 639 |         "object": "chat.completion",
 640 |         "usage": {
 641 |           "completion_tokens": 52,
 642 |           "prompt_tokens": 8,
 643 |           "total_tokens": 60
 644 |         }
 645 |       },
 646 |       "official": {
 647 |         "id": "chatcmpl-B5ny062ZzdKkkR3lDxsTvFtvcjNXr",
 648 |         "object": "chat.completion",
 649 |         "created": 1740723412,
 650 |         "model": "gpt-4o-2024-11-20",
 651 |         "choices": [
 652 |           {
 653 |             "index": 0,
 654 |             "message": {
 655 |               "role": "assistant",
 656 |               "content": "这是一张谷歌（Google）标志的图片。",
 657 |               "refusal": null
 658 |             },
 659 |             "logprobs": null,
 660 |             "finish_reason": "stop"
 661 |           }
 662 |         ],
 663 |         "usage": {
 664 |           "prompt_tokens": 452,
 665 |           "completion_tokens": 14,
 666 |           "total_tokens": 466,
 667 |           "prompt_tokens_details": {
 668 |             "cached_tokens": 0,
 669 |             "audio_tokens": 0
 670 |           },
 671 |           "completion_tokens_details": {
 672 |             "reasoning_tokens": 0,
 673 |             "audio_tokens": 0,
 674 |             "accepted_prediction_tokens": 0,
 675 |             "rejected_prediction_tokens": 0
 676 |           }
 677 |         },
 678 |         "service_tier": "default",
 679 |         "system_fingerprint": "fp_4dd9e4e381"
 680 |       }
 681 |     },
 682 |     "stop": {
 683 |       "reverse": {
 684 |         "id": "chatcmpl-bb5034c3-2e24-4ed1-b733-12e66505ce1a",
 685 |         "object": "chat.completion",
 686 |         "created": 1740734955,
 687 |         "system_fingerprint": "fp_d5139097a00d",
 688 |         "choices": [
 689 |           {
 690 |             "index": 0,
 691 |             "message": {
 692 |               "role": "assistant",
 693 |               "content": "Sure! Here you go:\n\n1, 2, 3, 4, 5, 6, 7, 8, 9, 10."
 694 |             },
 695 |             "finish_reason": "stop"
 696 |           }
 697 |         ],
 698 |         "usage": {
 699 |           "prompt_tokens": 7,
 700 |           "completion_tokens": 35,
 701 |           "total_tokens": 42
 702 |         }
 703 |       },
 704 |       "official": {
 705 |         "id": "chatcmpl-B5cNThpGsSY90qFeJYEs1SkG6D073",
 706 |         "object": "chat.completion",
 707 |         "created": 1740678863,
 708 |         "model": "gpt-4o-2024-08-06",
 709 |         "choices": [
 710 |           {
 711 |             "index": 0,
 712 |             "message": {
 713 |               "role": "assistant",
 714 |               "content": "Certainly! Here you go:\n\n1, 2, 3, ",
 715 |               "refusal": null
 716 |             },
 717 |             "logprobs": null,
 718 |             "finish_reason": "stop"
 719 |           }
 720 |         ],
 721 |         "usage": {
 722 |           "prompt_tokens": 14,
 723 |           "completion_tokens": 15,
 724 |           "total_tokens": 29,
 725 |           "prompt_tokens_details": {
 726 |             "cached_tokens": 0,
 727 |             "audio_tokens": 0
 728 |           },
 729 |           "completion_tokens_details": {
 730 |             "reasoning_tokens": 0,
 731 |             "audio_tokens": 0,
 732 |             "accepted_prediction_tokens": 0,
 733 |             "rejected_prediction_tokens": 0
 734 |           }
 735 |         },
 736 |         "service_tier": "default",
 737 |         "system_fingerprint": "fp_b491d0b9e9"
 738 |       }
 739 |     },
 740 |     "reasoning_max_tokens": {
 741 |       "reverse": {
 742 |         "id": "1c24e05b-b008-435a-9a93-6a14218b13c0",
 743 |         "object": "chat.completion",
 744 |         "created": 1743440167261,
 745 |         "model": "o3-mini",
 746 |         "choices": [
 747 |           {
 748 |             "index": 0,
 749 |             "message": {
 750 |               "role": "assistant",
 751 |               "content": "## Thought:\n\n\n\nReasoned about output repetition for a few seconds\n\n\n\n## Answer:\n\ndeepseek number 1  \ndeepseek number 1  \ndeepseek number 1  \ndeepseek number 1  \ndeepseek number 1"
 752 |             },
 753 |             "finish_reason": "stop"
 754 |           }
 755 |         ]
 756 |       },
 757 |       "official": {
 758 |         "id": "gen-1743440107-lNK2x8KVMirOwk3NYukG",
 759 |         "provider": "OpenAI",
 760 |         "model": "openai/o3-mini",
 761 |         "object": "chat.completion",
 762 |         "created": 1743440107,
 763 |         "choices": [
 764 |           {
 765 |             "logprobs": null,
 766 |             "finish_reason": "length",
 767 |             "native_finish_reason": "length",
 768 |             "index": 0,
 769 |             "message": {
 770 |               "role": "assistant",
 771 |               "content": "",
 772 |               "refusal": null
 773 |             }
 774 |           }
 775 |         ],
 776 |         "system_fingerprint": "fp_617f206dd9",
 777 |         "usage": {
 778 |           "prompt_tokens": 17,
 779 |           "completion_tokens": 10,
 780 |           "total_tokens": 27,
 781 |           "prompt_tokens_details": {
 782 |             "cached_tokens": 0
 783 |           },
 784 |           "completion_tokens_details": {
 785 |             "reasoning_tokens": 10
 786 |           }
 787 |         }
 788 |       }
 789 |     }
 790 |   },
 791 |   "claude": {
 792 |     "max_tokens": {
 793 |       "reverse": {
 794 |         "id": "chatcmpl-ac3e1f9d84a441e189f463d9e373b",
 795 |         "object": "chat.completion",
 796 |         "created": 1740740434,
 797 |         "model": "claude-3-5-sonnet-20241022",
 798 |         "system_fingerprint": "fp_a24b4d720c",
 799 |         "choices": [
 800 |           {
 801 |             "index": 0,
 802 |             "message": {
 803 |               "role": "assistant",
 804 |               "content": "你可以通过简单的循环来实现这个输出。你可以想象一个循环结构，重复输出字符串 'deepseek number 1' 五次。每次循环都会打印这个字符串，直到达到五次为止。"
 805 |             },
 806 |             "finish_reason": "stop"
 807 |           }
 808 |         ],
 809 |         "usage": {
 810 |           "prompt_tokens": 8,
 811 |           "completion_tokens": 23,
 812 |           "total_tokens": 31
 813 |         }
 814 |       },
 815 |       "official": {
 816 |         "id": "chatcmpl-beiu3LcIeCg5JyJwebH650FToElv6",
 817 |         "object": "chat.completion",
 818 |         "created": 1740740337,
 819 |         "model": "claude-3-5-sonnet-20241022",
 820 |         "choices": [
 821 |           {
 822 |             "index": 0,
 823 |             "message": {
 824 |               "role": "assistant",
 825 |               "content": "Here are 5 repetitions of '",
 826 |               "refusal": null
 827 |             },
 828 |             "logprobs": null,
 829 |             "finish_reason": "stop"
 830 |           }
 831 |         ],
 832 |         "usage": {
 833 |           "prompt_tokens": 22,
 834 |           "completion_tokens": 10,
 835 |           "total_tokens": 32
 836 |         }
 837 |       }
 838 |     },
 839 |     "stop": {
 840 |       "reverse": {
 841 |         "id": "chatcmpl-62b7e2cfd5a345b9b572182c8670b",
 842 |         "object": "chat.completion",
 843 |         "created": 1740740227,
 844 |         "model": "claude-3-5-sonnet-20241022",
 845 |         "system_fingerprint": "fp_a24b4d720c",
 846 |         "choices": [
 847 |           {
 848 |             "index": 0,
 849 |             "message": {
 850 |               "role": "assistant",
 851 |               "content": "Here is the count from 1 to 10 using Arabic numerals:\n\n1, 2, 3, 4, 5, 6, 7, 8, 9, 10"
 852 |             },
 853 |             "finish_reason": "stop"
 854 |           }
 855 |         ],
 856 |         "usage": {
 857 |           "prompt_tokens": 10,
 858 |           "completion_tokens": 23,
 859 |           "total_tokens": 33
 860 |         }
 861 |       },
 862 |       "official": {
 863 |         "id": "chatcmpl-EG0noygvQICNxMc7oTJobFt9r09ZP",
 864 |         "object": "chat.completion",
 865 |         "created": 1740740283,
 866 |         "model": "claude-3-5-sonnet-20241022",
 867 |         "choices": [
 868 |           {
 869 |             "index": 0,
 870 |             "message": {
 871 |               "role": "assistant",
 872 |               "content": "1\n2\n3\n",
 873 |               "refusal": null
 874 |             },
 875 |             "logprobs": null,
 876 |             "finish_reason": "stop"
 877 |           }
 878 |         ],
 879 |         "usage": {
 880 |           "prompt_tokens": 20,
 881 |           "completion_tokens": 8,
 882 |           "total_tokens": 28
 883 |         }
 884 |       }
 885 |     },
 886 |     "function_call": {
 887 |       "reverse": {
 888 |         "id": "chatcmpl-cf84c54e0443436fa0910dd982772",
 889 |         "object": "chat.completion",
 890 |         "created": 1740740119,
 891 |         "model": "claude-3-5-sonnet-20241022",
 892 |         "system_fingerprint": "fp_a24b4d720c",
 893 |         "choices": [
 894 |           {
 895 |             "index": 0,
 896 |             "message": {
 897 |               "role": "assistant",
 898 |               "content": "I'm unable to provide real-time information such as current weather conditions. I recommend checking a reliable weather website or app for the latest updates on the weather in Boston."
 899 |             },
 900 |             "finish_reason": "stop"
 901 |           }
 902 |         ],
 903 |         "usage": {
 904 |           "prompt_tokens": 10,
 905 |           "completion_tokens": 48,
 906 |           "total_tokens": 58
 907 |         }
 908 |       },
 909 |       "official": {
 910 |         "id": "msg_bdrk_01QTe5d7DHQADTBXHDNAVrEH",
 911 |         "model": "claude-3-5-sonnet-20241022",
 912 |         "object": "chat.completion",
 913 |         "created": 1740739157,
 914 |         "choices": [
 915 |           {
 916 |             "index": 0,
 917 |             "message": {
 918 |               "role": "assistant",
 919 |               "content": "I'll help you check the current weather in Boston. Let me get that information for you.",
 920 |               "tool_calls": [
 921 |                 {
 922 |                   "id": "toolu_bdrk_01LxMrXghyBxrUVhroBzGpJV",
 923 |                   "type": "function",
 924 |                   "function": {
 925 |                     "name": "get_current_weather",
 926 |                     "arguments": "{\"location\":\"Boston, MA\"}"
 927 |                   }
 928 |                 }
 929 |               ]
 930 |             },
 931 |             "finish_reason": "tool_use"
 932 |           }
 933 |         ],
 934 |         "usage": {
 935 |           "prompt_tokens": 426,
 936 |           "completion_tokens": 77,
 937 |           "total_tokens": 503
 938 |         }
 939 |       }
 940 |     },
 941 |     "reasoning_stop": {
 942 |       "reverse": {
 943 |         "id": "chatcmpl-08c0e27865f14316b8102656455ea53d",
 944 |         "object": "chat.completion",
 945 |         "created": 1743445861,
 946 |         "model": "claude-3-7-sonnet-20250219-thinking",
 947 |         "choices": [
 948 |           {
 949 |             "index": 0,
 950 |             "message": {
 951 |               "role": "assistant",
 952 |               "content": "以下是从1到10的计数：\n\n1\n2\n3\n4\n5\n6\n7\n8\n9\n10"
 953 |             }
 954 |           }
 955 |         ]
 956 |       },
 957 |       "official": {
 958 |         "id": "gen-1743445740-2bw4LAyVVgNwDbAG6mNU",
 959 |         "provider": "Google",
 960 |         "model": "anthropic/claude-3.7-sonnet",
 961 |         "object": "chat.completion",
 962 |         "created": 1743445740,
 963 |         "choices": [
 964 |           {
 965 |             "logprobs": null,
 966 |             "finish_reason": "stop",
 967 |             "native_finish_reason": "stop",
 968 |             "index": 0,
 969 |             "message": {
 970 |               "role": "assistant",
 971 |               "content": "Here's counting from 1 to 10:\n\n1\n2\n",
 972 |               "refusal": null,
 973 |               "reasoning": "The human is asking me to count from 1 to 10. This is a simple task where I need to list the numbers in ascending order from 1 to 10.\n\n1\n2\n3\n4\n5\n6\n7\n8\n9\n10\n\nThat's the complete count from 1 to 10."
 974 |             }
 975 |           }
 976 |         ],
 977 |         "usage": {
 978 |           "prompt_tokens": 43,
 979 |           "completion_tokens": 100,
 980 |           "total_tokens": 143
 981 |         }
 982 |       }
 983 |     }
 984 |   },
 985 |   "gemini": {
 986 |     "max_tokens": {
 987 |       "reverse": {
 988 |         "id": "chatcmpl-zsFrobjxCsqpeuCNUfJaFBK4OXMTr",
 989 |         "object": "chat.completion",
 990 |         "created": 1740740838,
 991 |         "model": "gemini-2.0-flash-thinking-exp",
 992 |         "choices": [
 993 |           {
 994 |             "index": 0,
 995 |             "message": {
 996 |               "role": "assistant",
 997 |               "content": "好的，没问题！\n\ndeepseek number 1 deepseek number 1 deepseek number 1 deepseek number 1 deepseek number 1\n\n希望这个输出符合你的要求！ 如果你还有其他指令，尽管告诉我。",
 998 |               "refusal": null
 999 |             },
1000 |             "logprobs": null,
1001 |             "finish_reason": "stop"
1002 |           }
1003 |         ],
1004 |         "usage": {
1005 |           "prompt_tokens": 0,
1006 |           "completion_tokens": 0,
1007 |           "total_tokens": 0,
1008 |           "prompt_tokens_details": {
1009 |             "cached_tokens": 0,
1010 |             "audio_tokens": 0
1011 |           },
1012 |           "completion_tokens_details": {
1013 |             "reasoning_tokens": 0,
1014 |             "audio_tokens": 0,
1015 |             "accepted_prediction_tokens": 0,
1016 |             "rejected_prediction_tokens": 0
1017 |           }
1018 |         },
1019 |         "system_fingerprint": "fp_4a61d2ebb8"
1020 |       },
1021 |       "official": {
1022 |         "id": "chatcmpl-0915722c518345f894d9aeb4efc77e28",
1023 |         "model": "gemini-2.0-flash",
1024 |         "object": "chat.completion",
1025 |         "created": 1740741172,
1026 |         "choices": [
1027 |           {
1028 |             "index": 0,
1029 |             "message": {
1030 |               "role": "assistant",
1031 |               "content": "deepseek number 1\ndeepseek number "
1032 |             },
1033 |             "finish_reason": "length"
1034 |           }
1035 |         ],
1036 |         "error": null,
1037 |         "usage": {
1038 |           "prompt_tokens": 16,
1039 |           "completion_tokens": 10,
1040 |           "total_tokens": 26,
1041 |           "prompt_tokens_details": {
1042 |             "cached_tokens": 0,
1043 |             "text_tokens": 0,
1044 |             "audio_tokens": 0,
1045 |             "image_tokens": 0
1046 |           },
1047 |           "completion_tokens_details": {
1048 |             "text_tokens": 0,
1049 |             "audio_tokens": 0
1050 |           }
1051 |         }
1052 |       }
1053 |     },
1054 |     "codeExecution": {
1055 |       "reverse": {
1056 |         "id": "chatcmpl-JVxi2NwSIJVkwoNzuozEjhVKiOQbU",
1057 |         "object": "chat.completion",
1058 |         "created": 1740742479,
1059 |         "model": "gemini-2.0-flash-thinking-exp",
1060 |         "choices": [
1061 |           {
1062 |             "index": 0,
1063 |             "message": {
1064 |               "role": "assistant",
1065 |               "content": "The sum of the first 50 prime numbers is a fascinating question! Let's embark on a journey to discover this sum. We'll need to identify those elusive prime numbers first, and then, with a bit of computational magic, we'll add them all up.\n\nHere's a little code snippet in Python to help us on our quest:\n\n```python\ndef is_prime(n):\n    if n <= 1:\n        return False\n    if n <= 3:\n        return True\n    if n % 2 == 0 or n % 3 == 0:\n        return False\n    i = 5\n    while i * i <= n:\n        if n % i == 0 or n % (i + 2) == 0:\n            return False\n        i += 6\n    return True\n\ndef sum_first_n_primes(n):\n    primes = []\n    num = 2\n    while len(primes) < n:\n        if is_prime(num):\n            primes.append(num)\n        num += 1\n    return sum(primes)\n\nnumber_of_primes = 50\nsum_of_primes = sum_first_n_primes(number_of_primes)\nprint(f\"The sum of the first {number_of_primes} prime numbers is: {sum_of_primes}\")\n\n```\n\nWhen we run this code, it diligently calculates and reveals that the sum of the first 50 prime numbers is **24133**.\n\nIsn't it amazing how numbers hold such secrets waiting to be uncovered?  Prime numbers, in particular, have intrigued mathematicians for centuries, and their distribution and properties continue to be a source of wonder and study.",
1066 |               "refusal": null
1067 |             },
1068 |             "logprobs": null,
1069 |             "finish_reason": "stop"
1070 |           }
1071 |         ],
1072 |         "usage": {
1073 |           "prompt_tokens": 0,
1074 |           "completion_tokens": 0,
1075 |           "total_tokens": 0,
1076 |           "prompt_tokens_details": {
1077 |             "cached_tokens": 0,
1078 |             "audio_tokens": 0
1079 |           },
1080 |           "completion_tokens_details": {
1081 |             "reasoning_tokens": 0,
1082 |             "audio_tokens": 0,
1083 |             "accepted_prediction_tokens": 0,
1084 |             "rejected_prediction_tokens": 0
1085 |           }
1086 |         },
1087 |         "system_fingerprint": "fp_a713964b2e"
1088 |       },
1089 |       "official": {
1090 |         "id": "chatcmpl-1b79cf65137f4601a498c5c95b701b64",
1091 |         "model": "gemini-2.0-flash",
1092 |         "object": "chat.completion",
1093 |         "created": 1740742508,
1094 |         "choices": [
1095 |           {
1096 |             "index": 0,
1097 |             "message": {
1098 |               "role": "assistant",
1099 |               "content": "Okay, I need to calculate the sum of the first 50 prime numbers. Here's my plan:\n\n1.  **Generate Prime Numbers:** Create a function to generate prime numbers. I will iterate through numbers, checking for primality.\n2.  **Store Prime Numbers:** Store the generated prime numbers in a list until I have 50 of them.\n3.  **Sum the Primes:** Calculate the sum of the prime numbers in the list.\n4.  **Output:** Present the sum.\n\nHere's the Python code:\n\n\n```PYTHON\ndef is_prime(n):\n    \"\"\"Check if a number is prime.\"\"\"\n    if n <= 1:\n        return False\n    if n <= 3:\n        return True\n    if n % 2 == 0 or n % 3 == 0:\n        return False\n    i = 5\n    while i * i <= n:\n        if n % i == 0 or n % (i + 2) == 0:\n            return False\n        i += 6\n    return True\n\nprimes = []\nnum = 2\nwhile len(primes) < 50:\n    if is_prime(num):\n        primes.append(num)\n    num += 1\n\nsum_of_primes = sum(primes)\nprint(f'{primes=}')\nprint(f'{sum_of_primes=}')\n\n```\n```output\nprimes=[2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89, 97, 101, 103, 107, 109, 113, 127, 131, 137, 139, 149, 151, 157, 163, 167, 173, 179, 181, 191, 193, 197, 199, 211, 223, 227, 229]\nsum_of_primes=5117\n\n```\nThe sum of the first 50 prime numbers is 5117.\n"
1100 |             },
1101 |             "finish_reason": "stop"
1102 |           }
1103 |         ],
1104 |         "error": null,
1105 |         "usage": {
1106 |           "prompt_tokens": 31,
1107 |           "completion_tokens": 560,
1108 |           "total_tokens": 1164,
1109 |           "prompt_tokens_details": {
1110 |             "cached_tokens": 0,
1111 |             "text_tokens": 0,
1112 |             "audio_tokens": 0,
1113 |             "image_tokens": 0
1114 |           },
1115 |           "completion_tokens_details": {
1116 |             "text_tokens": 0,
1117 |             "audio_tokens": 0
1118 |           }
1119 |         }
1120 |       }
1121 |     },
1122 |     "googleSearch": {
1123 |       "reverse": {
1124 |         "id": "chatcmpl-nOOGow6EqWOsXskZmEMEqxvgfyU7L",
1125 |         "object": "chat.completion",
1126 |         "created": 1740742300,
1127 |         "model": "gemini-2.0-flash-thinking-exp",
1128 |         "choices": [
1129 |           {
1130 |             "index": 0,
1131 |             "message": {
1132 |               "role": "assistant",
1133 |               "content": "今天是一月二十六号。",
1134 |               "refusal": null
1135 |             },
1136 |             "logprobs": null,
1137 |             "finish_reason": "stop"
1138 |           }
1139 |         ],
1140 |         "usage": {
1141 |           "prompt_tokens": 0,
1142 |           "completion_tokens": 0,
1143 |           "total_tokens": 0,
1144 |           "prompt_tokens_details": {
1145 |             "cached_tokens": 0,
1146 |             "audio_tokens": 0
1147 |           },
1148 |           "completion_tokens_details": {
1149 |             "reasoning_tokens": 0,
1150 |             "audio_tokens": 0,
1151 |             "accepted_prediction_tokens": 0,
1152 |             "rejected_prediction_tokens": 0
1153 |           }
1154 |         },
1155 |         "system_fingerprint": "fp_082af1d649"
1156 |       },
1157 |       "official": {
1158 |         "id": "chatcmpl-ee821b0f754d432bac99dddaf0bcc88e",
1159 |         "model": "gemini-2.0-flash",
1160 |         "object": "chat.completion",
1161 |         "created": 1740742342,
1162 |         "choices": [
1163 |           {
1164 |             "index": 0,
1165 |             "message": {
1166 |               "role": "assistant",
1167 |               "content": "今天是2025年2月28日。\n"
1168 |             },
1169 |             "finish_reason": "stop"
1170 |           }
1171 |         ],
1172 |         "error": null,
1173 |         "usage": {
1174 |           "prompt_tokens": 7,
1175 |           "completion_tokens": 14,
1176 |           "total_tokens": 21,
1177 |           "prompt_tokens_details": {
1178 |             "cached_tokens": 0,
1179 |             "text_tokens": 0,
1180 |             "audio_tokens": 0,
1181 |             "image_tokens": 0
1182 |           },
1183 |           "completion_tokens_details": {
1184 |             "text_tokens": 0,
1185 |             "audio_tokens": 0
1186 |           }
1187 |         }
1188 |       }
1189 |     },
1190 |     "response_format": {
1191 |       "reverse": {
1192 |         "id": "chatcmpl-Jj0K7EANLPteO602hziq673IYnNLC",
1193 |         "object": "chat.completion",
1194 |         "created": 1740741510,
1195 |         "model": "gemini-2.0-flash-thinking-exp",
1196 |         "choices": [
1197 |           {
1198 |             "index": 0,
1199 |             "message": {
1200 |               "role": "assistant",
1201 |               "content": "当然！ 让我给你一个名字和年龄：\n\n名字： 李明 (Lǐ Míng)\n年龄： 25 岁",
1202 |               "refusal": null
1203 |             },
1204 |             "logprobs": null,
1205 |             "finish_reason": "stop"
1206 |           }
1207 |         ],
1208 |         "usage": {
1209 |           "prompt_tokens": 0,
1210 |           "completion_tokens": 0,
1211 |           "total_tokens": 0,
1212 |           "prompt_tokens_details": {
1213 |             "cached_tokens": 0,
1214 |             "audio_tokens": 0
1215 |           },
1216 |           "completion_tokens_details": {
1217 |             "reasoning_tokens": 0,
1218 |             "audio_tokens": 0,
1219 |             "accepted_prediction_tokens": 0,
1220 |             "rejected_prediction_tokens": 0
1221 |           }
1222 |         },
1223 |         "system_fingerprint": "fp_0cb7f95a34"
1224 |       },
1225 |       "official": {
1226 |         "id": "chatcmpl-9585ed9d59aa42499b5adf73473b7382",
1227 |         "model": "gemini-2.0-flash",
1228 |         "object": "chat.completion",
1229 |         "created": 1740741918,
1230 |         "choices": [
1231 |           {
1232 |             "index": 0,
1233 |             "message": {
1234 |               "role": "assistant",
1235 |               "content": "{\n  \"age\": 30,\n  \"name\": \"李明\"\n}"
1236 |             },
1237 |             "finish_reason": "stop"
1238 |           }
1239 |         ],
1240 |         "error": null,
1241 |         "usage": {
1242 |           "prompt_tokens": 10,
1243 |           "completion_tokens": 21,
1244 |           "total_tokens": 31,
1245 |           "prompt_tokens_details": {
1246 |             "cached_tokens": 0,
1247 |             "text_tokens": 0,
1248 |             "audio_tokens": 0,
1249 |             "image_tokens": 0
1250 |           },
1251 |           "completion_tokens_details": {
1252 |             "text_tokens": 0,
1253 |             "audio_tokens": 0
1254 |           }
1255 |         }
1256 |       }
1257 |     }
1258 |   }
1259 | }


--------------------------------------------------------------------------------
/src/data/prompts.json:
--------------------------------------------------------------------------------
  1 | {
  2 |   "openai": {
  3 |     "image_url": {
  4 |       "content": [
  5 |         {
  6 |           "type": "text",
  7 |           "text": "这是一张什么图片"
  8 |         },
  9 |         {
 10 |           "type": "image_url",
 11 |           "image_url": {
 12 |             "url": "https://www.google.com/images/branding/googlelogo/1x/googlelogo_light_color_272x92dp.png"
 13 |           }
 14 |         }
 15 |       ],
 16 |       "params": {
 17 |         "max_tokens": 100
 18 |       }
 19 |     },
 20 |     "max_tokens": {
 21 |       "content": "输出 'deepseek number 1' 5次",
 22 |       "params": {
 23 |         "max_tokens": 10
 24 |       }
 25 |     },
 26 |     "reasoning_max_tokens": {
 27 |       "content": "输出 'deepseek number 1' 5次",
 28 |       "params": {
 29 |         "max_tokens": 10
 30 |       }
 31 |     },
 32 |     "logprobs": {
 33 |       "content": "Hello!",
 34 |       "params": {
 35 |         "logprobs": true,
 36 |         "top_logprobs": 2
 37 |       }
 38 |     },
 39 |     "n": {
 40 |       "content": "Hello!",
 41 |       "params": {
 42 |         "n": 2
 43 |       }
 44 |     },
 45 |     "stop": {
 46 |       "content": "Count from 1 to 10",
 47 |       "params": {
 48 |         "stop": [
 49 |           "4"
 50 |         ]
 51 |       }
 52 |     },
 53 |     "function_call": {
 54 |       "content": "Whats the weather like in Boston today?",
 55 |       "params": {
 56 |         "tools": [
 57 |           {
 58 |             "type": "function",
 59 |             "function": {
 60 |               "name": "get_current_weather",
 61 |               "description": "Get the current weather in a given location",
 62 |               "parameters": {
 63 |                 "type": "object",
 64 |                 "properties": {
 65 |                   "location": {
 66 |                     "type": "string",
 67 |                     "description": "The city and state, e.g. San Francisco, CA"
 68 |                   },
 69 |                   "unit": {
 70 |                     "type": "string",
 71 |                     "enum": [
 72 |                       "celsius",
 73 |                       "fahrenheit"
 74 |                     ]
 75 |                   }
 76 |                 },
 77 |                 "required": [
 78 |                   "location"
 79 |                 ]
 80 |               }
 81 |             }
 82 |           }
 83 |         ],
 84 |         "tool_choice": "auto"
 85 |       }
 86 |     },
 87 |     "response_format": {
 88 |       "content": "给我一个名字和年龄吧",
 89 |       "params": {
 90 |         "response_format": {
 91 |           "type": "json_schema",
 92 |           "json_schema": {
 93 |             "name": "user_info",
 94 |             "schema": {
 95 |               "type": "object",
 96 |               "properties": {
 97 |                 "name": {
 98 |                   "type": "string"
 99 |                 },
100 |                 "age": {
101 |                   "type": "number"
102 |                 }
103 |               },
104 |               "required": [
105 |                 "name",
106 |                 "age"
107 |               ],
108 |               "additionalProperties": false
109 |             },
110 |             "strict": true
111 |           }
112 |         }
113 |       }
114 |     }
115 |   },
116 |   "claude": {
117 |     "max_tokens": {
118 |       "content": "输出 'deepseek number 1' 5次",
119 |       "params": {
120 |         "max_tokens": 10
121 |       }
122 |     },
123 |     "reasoning_stop": {
124 |       "content": "Count from 1 to 10 with Arabic numerals",
125 |       "params": {
126 |         "stop": ["3"]
127 |       }
128 |     },
129 |     "stop": {
130 |       "content": "Count from 1 to 10 with Arabic numerals",
131 |       "params": {
132 |         "stop": [
133 |           "4"
134 |         ]
135 |       }
136 |     },
137 |     "function_call": {
138 |       "content": "Whats the weather like in Boston today?",
139 |       "params": {
140 |         "tools": [
141 |           {
142 |             "type": "function",
143 |             "function": {
144 |               "name": "get_current_weather",
145 |               "description": "Get the current weather in a given location",
146 |               "parameters": {
147 |                 "type": "object",
148 |                 "properties": {
149 |                   "location": {
150 |                     "type": "string",
151 |                     "description": "The city and state, e.g. San Francisco, CA"
152 |                   },
153 |                   "unit": {
154 |                     "type": "string",
155 |                     "enum": [
156 |                       "celsius",
157 |                       "fahrenheit"
158 |                     ]
159 |                   }
160 |                 },
161 |                 "required": [
162 |                   "location"
163 |                 ]
164 |               }
165 |             }
166 |           }
167 |         ],
168 |         "tool_choice": "auto"
169 |       }
170 |     }
171 |   },
172 |   "gemini": {
173 |     "max_tokens": {
174 |       "content": "输出 'deepseek number 1' 5次",
175 |       "params": {
176 |         "max_tokens": 10
177 |       }
178 |     },
179 |     "codeExecution": {
180 |       "content": "What is the sum of the first 50 prime numbers? Generate and run code for the calculation, and make sure you get all 50.",
181 |       "params": {
182 |         "tools": [
183 |           {
184 |             "type": "function",
185 |             "function": {
186 |               "name": "codeExecution"
187 |             }
188 |           }
189 |         ]
190 |       }
191 |     },
192 |     "googleSearch": {
193 |       "content": "今天是几月几号.",
194 |       "params": {
195 |         "tools": [
196 |           {
197 |             "type": "function",
198 |             "function": {
199 |               "name": "googleSearch"
200 |             }
201 |           }
202 |         ]
203 |       }
204 |     },
205 |     "response_format": {
206 |       "content": "给我一个名字和年龄吧",
207 |       "params": {
208 |         "response_format": {
209 |           "type": "json_schema",
210 |           "json_schema": {
211 |             "name": "user_info",
212 |             "schema": {
213 |               "type": "object",
214 |               "properties": {
215 |                 "name": {
216 |                   "type": "string"
217 |                 },
218 |                 "age": {
219 |                   "type": "number"
220 |                 }
221 |               },
222 |               "required": [
223 |                 "name",
224 |                 "age"
225 |               ],
226 |               "additionalProperties": false
227 |             },
228 |             "strict": true
229 |           }
230 |         }
231 |       }
232 |     }
233 |   }
234 | }


--------------------------------------------------------------------------------
/src/index.css:
--------------------------------------------------------------------------------
 1 | @tailwind base;
 2 | @tailwind components;
 3 | @tailwind utilities;
 4 | 
 5 | :root {
 6 |   --primary-color: #4f46e5;
 7 |   --primary-hover: #4338ca;
 8 |   --secondary-color: #0ea5e9;
 9 |   --secondary-hover: #0284c7;
10 |   --bg-color: #f8fafc;
11 |   --card-bg: #ffffff;
12 |   --text-color: #1e293b;
13 |   --text-light: #64748b;
14 |   --border-color: #e2e8f0;
15 |   --shadow-sm: 0 1px 2px 0 rgba(0, 0, 0, 0.05);
16 |   --shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06);
17 |   --shadow-md: 0 10px 15px -3px rgba(0, 0, 0, 0.1), 0 4px 6px -2px rgba(0, 0, 0, 0.05);
18 | }
19 | 
20 | body {
21 |   background-color: var(--bg-color);
22 |   color: var(--text-color);
23 | }
24 | 
25 | @layer components {
26 |   .btn-primary {
27 |     @apply bg-indigo-600 text-white px-4 py-2 rounded-md hover:bg-indigo-700 focus:outline-none focus:ring-2 focus:ring-indigo-500 focus:ring-offset-2 transition-all duration-200;
28 |   }
29 |   
30 |   .btn-secondary {
31 |     @apply bg-sky-500 text-white px-4 py-2 rounded-md hover:bg-sky-600 focus:outline-none focus:ring-2 focus:ring-sky-400 focus:ring-offset-2 transition-all duration-200;
32 |   }
33 |   
34 |   .card {
35 |     @apply bg-white rounded-xl shadow-md border border-gray-100 overflow-hidden transition-shadow duration-300 hover:shadow-lg;
36 |   }
37 |   
38 |   .input-field {
39 |     @apply w-full border border-gray-300 rounded-md px-3 py-2 focus:outline-none focus:ring-2 focus:ring-indigo-500 focus:border-indigo-500 transition-all duration-200;
40 |   }
41 |   
42 |   .select-field {
43 |     @apply w-full border border-gray-300 rounded-md px-3 py-2 focus:outline-none focus:ring-2 focus:ring-indigo-500 focus:border-indigo-500 transition-all duration-200;
44 |   }
45 | }


--------------------------------------------------------------------------------
/src/main.js:
--------------------------------------------------------------------------------
 1 | import { createApp } from 'vue'
 2 | import App from './App.vue'
 3 | import { createRouter, createWebHistory } from 'vue-router'
 4 | import { createI18n } from 'vue-i18n'
 5 | import './index.css'
 6 | 
 7 | // 路由配置
 8 | const router = createRouter({
 9 |   history: createWebHistory(),
10 |   routes: [
11 |     {
12 |       path: '/',
13 |       component: () => import('./views/Home.vue')
14 |     }
15 |   ]
16 | })
17 | 
18 | // i18n配置
19 | const i18n = createI18n({
20 |   legacy: false,
21 |   locale: 'zh-CN',
22 |   messages: {
23 |     'zh-CN': {
24 |       message: {
25 |         title: '模型逆向与否检测',
26 |         startCheck: '开始检测',
27 |         modelProvider: '模型提供商',
28 |         testParams: '检测参数',
29 |         apiBaseUrl: 'API Base URL',
30 |         apiKey: 'API Key',
31 |         apiKeyPlaceholder: '请输入您的API密钥',
32 |         apiBaseUrlPlaceholder: '例如：api.openai.com（无需添加https://或/v1）',
33 |         model: '待检测模型',
34 |         modelPlaceholder: '例如：gpt-4o-2024-11-20, gpt-3.5-turbo',
35 |         reverseExample: '逆向接口返回示例',
36 |         officialExample: '官方接口返回示例',
37 |         requestResult: '请求结果',
38 |         showBriefResponse: '显示简要响应',
39 |         showFullResponse: '显示完整响应',
40 |         apiDocs: 'API文档链接',
41 |       }
42 |     }
43 |   }
44 | })
45 | 
46 | const app = createApp(App)
47 | 
48 | app.use(router)
49 | app.use(i18n)
50 | 
51 | app.mount('#app')


--------------------------------------------------------------------------------
/src/views/Home.vue:
--------------------------------------------------------------------------------
  1 | 
252 | 
253 | 
460 | 
461 | 


--------------------------------------------------------------------------------
/tailwind.config.js:
--------------------------------------------------------------------------------
 1 | /** @type {import('tailwindcss').Config} */
 2 | export default {
 3 |   content: [
 4 |     "./index.html",
 5 |     "./src/**/*.{vue,js,ts,jsx,tsx}",
 6 |   ],
 7 |   theme: {
 8 |     extend: {},
 9 |   },
10 |   plugins: [],
11 | }


--------------------------------------------------------------------------------
/vite.config.js:
--------------------------------------------------------------------------------
 1 | import { defineConfig } from 'vite'
 2 | import vue from '@vitejs/plugin-vue'
 3 | 
 4 | export default defineConfig({
 5 |   plugins: [vue()],
 6 |   server: {
 7 |     port: 3000,
 8 |     proxy: {
 9 |       '/api': {
10 |         target: 'http://localhost:3000',
11 |         changeOrigin: true,
12 |         rewrite: (path) => path.replace(/^\/api/, '')
13 |       }
14 |     }
15 |   }
16 | })


--------------------------------------------------------------------------------