> ## Documentation Index > Fetch the complete documentation index at: https://docs.socialvision.tisyk.xyz/llms.txt > Use this file to discover all available pages before exploring further. # Python SDK 与代码集成指南 > 精选 Python 基础对话、流式输出、函数调用与图片识别调用合集 ## Python基础对话 [您可以通过任何语言的 HTTP 请求、我们的官方 Python 绑定、我们的官方 Node.js 库或社区维护的库](https://platform.openai.com/docs/libraries/community-libraries)与 API 交互。 要安装官方 Python 绑定，请运行以下命令： ```bash theme={null} pip install openai ``` ```python theme={null} key='sk-xxxx' from openai import OpenAI client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key=key ) response = client.chat.completions.create( model="gpt-4o", messages=[ {"role": "user", "content": "你好?"}, ], timeout=100, ) print(response) ``` *** ## python 连续对话 ```python theme={null} import openai import time # 打印openai的版本 print(openai.__version__) client = openai.OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key="sk-xxx" ) messages = [ {"role": "system", "content": "You are a helpful assistant."} ] while True: user_input = input("You: ") messages.append({"role": "user", "content": user_input}) response = client.chat.completions.create( model="gpt-4o", messages=messages, temperature=1 ) assistant_response = response.choices[0].message.content print(f"Assistant: {assistant_response}") messages.append({"role": "assistant", "content": assistant_response}) time.sleep(1) ``` *** ## python llama\_index 配置 ```python theme={null} from llama_index.llms.openai import OpenAI llm = OpenAI( model="gpt-3.5-turbo", api_key="sk-xxxxx", api_base="https://socialvision.tisyk.xyz/v1") ret=llm.complete("Paul Graham is ") print(ret) ``` *** ## python 调用DALL·E ## 功能概述 API 提供三种主要功能： 1. 🎨 文本生成图像 (DALL·E 3 和 DALL·E 2) 2. ✏️ 图像编辑 (仅 DALL·E 2) 3. 🔄 图像变体生成 (仅 DALL·E 2) ## 1. 图像生成 ### 基础使用 ```python theme={null} from openai import OpenAI client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key=key ) response = client.images.generate( model="dall-e-3", prompt="a white siamese cat", size="1024x1024", quality="standard", n=1, ) image_url = response.data[0].url ``` ### 参数说明 * **尺寸选项**：1024x1024、1024x1792、1792x1024 * **质量选项**：standard(默认)、hd(DALL·E 3专属) * **数量限制**： * DALL·E 3：单次1张 * DALL·E 2：单次最多10张 ## 2. 图像编辑 (DALL·E 2) ### 使用示例 ```python theme={null} from openai import OpenAI client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key=key ) response = client.images.edit(( model="dall-e-2", image=open("sunlit_lounge.png", "rb"), mask=open("mask.png", "rb"), prompt="A sunlit indoor lounge area with a pool containing a flamingo", n=1, size="1024x1024" ) image_url = response.data[0].url ``` ### 要求说明 * 📝 图片和蒙版必须为PNG格式 * 📏 必须为方形图片 * 💾 文件大小\<4MB * ⚖️ 图片和蒙版尺寸必须相同 ## 3. 图像变体生成 (DALL·E 2) ### 使用示例 ```python theme={null} from openai import OpenAI client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key=key ) response = client.images.create_variation( model="dall-e-2", image=open("corgi_and_cat_paw.png", "rb"), n=1, size="1024x1024" ) image_url = response.data[0].url ``` ### 技术要求 * 📝 PNG格式 * 📏 方形图片 * 💾 文件大小\<4MB ## 提示技巧 ### DALL·E 3 特性 * 自动优化提示 * 可通过特殊指令控制提示优化： ``` I NEED to test how the tool works with extremely simple prompts. DO NOT add any detail, just use it AS-IS: ``` ## 注意事项 ### 内容审核 * 严格遵循内容政策 * 违规内容将返回错误 ### 图片URL * 有效期为1小时 * 可选择Base64格式返回 *** ## python 使用Embeddings 向量化 ## 概述 ### 新模型发布 * **text-embedding-3-small** * **text-embedding-3-large** 特点：更低成本、更好的多语言性能、可控制维度 ### 主要应用场景 * 🔍 搜索（相关性排序） * 📊 聚类（相似性分组） * 👍 推荐系统 * ⚠️ 异常检测 * 📈 多样性分析 * 🏷️ 文本分类 ## 基础使用 ### 获取嵌入向量 ```python theme={null} from openai import OpenAI client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key=key ) response = client.embeddings.create( input="Your text string goes here", model="text-embedding-3-small" ) print(response.data[0].embedding) ``` ### 响应格式 ```json theme={null} { "object": "list", "data": [ { "object": "embedding", "index": 0, "embedding": [ -0.006929283495992422, -0.005336422007530928, // ... 更多数值 ], } ], "model": "text-embedding-3-small", "usage": { "prompt_tokens": 5, "total_tokens": 5 } } ``` ## 模型对比 | 模型 | 每美元页面数 | MTEB性能评估 | 最大输入 | | - | - | - | - | | text-embedding-3-small | 62,500 | 62.3% | 8191 | | text-embedding-3-large | 9,615 | 64.6% | 8191 | | text-embedding-ada-002 | 12,500 | 61.0% | 8191 | ## 实际应用示例 ### 处理评论数据 ```python theme={null} from openai import OpenAI client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key=key ) def get_embedding(text, model="text-embedding-3-small"): text = text.replace("\n", " ") return client.embeddings.create(input = [text], model=model).data[0].embedding # 处理数据框 df['ada_embedding'] = df.combined.apply(lambda x: get_embedding(x, model='text-embedding-3-small')) df.to_csv('output/embedded_1k_reviews.csv', index=False) # 加载保存的嵌入 import pandas as pd import numpy as np df = pd.read_csv('output/embedded_1k_reviews.csv') df['ada_embedding'] = df.ada_embedding.apply(eval).apply(np.array) ``` ## 技术细节 ### 维度说明 * text-embedding-3-small: 默认1536维 * text-embedding-3-large: 默认3072维 * 可通过 Dimensions 参数调整维度 ### 注意事项 * 计费基于输入令牌数 * 每页约800个令牌 * 所有模型最大输入均为8191个令牌 *** ## python openai官方库（使用AutoGPT，langchain等） ## 方法一：直接配置 OpenAI ```python theme={null} import openai # 设置 API 基础地址和密钥 openai.api_base = "https://socialvision.tisyk.xyz/v1" openai.api_key = "sk-xxxxxxxxx" ``` ## 方法二：环境变量配置 > 如果方法一不起作用，可以尝试此方法 需要设置以下环境变量： ```bash theme={null} OPENAI_API_BASE=https://socialvision.tisyk.xyz/v1 OPENAI_API_KEY=sk-xxxxx ``` > **注意**：修改环境变量后如不起作用，请重启系统 ## 方法三：使用 OpenAI 客户端 ```python theme={null} from openai import OpenAI # 初始化客户端 client = OpenAI( base_url="https://socialvision.tisyk.xyz/v1", api_key='您的API KEY', timeout=120 ) # 创建聊天完成 response = client.chat.completions.create( model="gpt-4o", messages=[ {"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "Who won the world series in 2020?"}, {"role": "assistant", "content": "The Los Angeles Dodgers won the World Series in 2020."}, {"role": "user", "content": "Where was it played?"} ] ) print(response) #response格式 response = client.responses.create( model="gpt-5", input="Write a one-sentence bedtime story about a unicorn." ) print(response.output_text) ``` *** ## python request 请求 流式输出demo ```python theme={null} import requests import json url = "https://socialvision.tisyk.xyz/v1/chat/completions" payload = json.dumps({ "model": "o1-preview", "messages": [ { "role": "user", "content": "写一个小情书" } ], "stream": True }) headers = { 'Accept': 'text/event-stream', 'Authorization': 'Bearer 你的key', 'Content-Type': 'application/json' } response = requests.request("POST", url, headers=headers, data=payload, stream=True) for line in response.iter_lines(): if line: # 跳过空行 line = line.decode('utf-8') if line.startswith('data: '): if line == 'data: [DONE]': break # 解析JSON数据 data = json.loads(line[6:]) # 去掉 "data: " 前缀 if 'choices' in data and len(data['choices']) > 0: delta = data['choices'][0].get('delta', {}) if 'content' in delta: print(delta['content'], end='', flush=True) ``` *** ## python 简单langchain 调用openai demo ## 识别链接格式图片 ```js theme={null} pip install langchain_openai pip install langchain ``` ```python theme={null} #新版本langchina from langchain_openai import ChatOpenAI llm = ChatOpenAI( openai_api_base="https://socialvision.tisyk.xyz/v1", openai_api_key="sk-xxxxx" ) res = llm.invoke("hello") print(res.content) #老版本langchina 通过设置环境变量的形式设置中转地址 $env:OPENAI_BASE_URL="https://socialvision.tisyk.xyz/v1" $env:OPENAI_API_KEY="sk-xxxxx" ``` *** This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.