Home/大语言模型/OpenAI API

OpenAI API

OpenAI 接口

Access GPT-4o, GPT-4, GPT-3.5, embeddings, DALL-E, Whisper, and TTS models via REST API.

通过 REST API 调用 GPT-4o、GPT-4、GPT-3.5、Embeddings、DALL-E、Whisper 和 TTS 模型。

Website

https://openai.com

Documentation

https://platform.openai.com/docs

Pricing

pay-as-you-go (per token)

Authentication

API Key (Bearer Token)

Rate Limit

10000 req/min (varies by tier)

Typical Latency

<500ms (GPT-4o)

Max Context

128K (GPT-4o)

Popularity

100/100

Endpoints

https://api.openai.com/v1/chat/completions

https://api.openai.com/v1/embeddings

https://api.openai.com/v1/images/generations

https://api.openai.com/v1/audio/transcriptions

SDKs

PythonNode.jsGoJavacurl

Key Features

◆GPT-4o, GPT-4, GPT-3.5 model family
◆Multimodal: vision, audio, text generation
◆Function calling & structured outputs
◆Embeddings API for semantic search
◆Fine-tuning for custom models

Code Examples

Chat Completionbash

curl https://api.openai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python: Chat Completionpython

from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)