ChatGPT Web Page to API interface.
- Bypass
Cloudflarechallenge - Login mode: no-login, email, google login
- High-speed streaming output
- Model switching, Dynamically display supported models
- Conversation support
- TTS support
Compatible with the ChatGPT API.
docker run --name llm-web-api --rm -it -p 5000:5000 adryfish/llm-web-apiSee detailed configuration below for environment variables.
services:
llm-web-api:
image: adryfish/llm-web-api
container_name: llm-web-api
ports:
- "5000:5000"
volumes:
# Browser data. Configure if you want to retain browser login information.
- ./data:/app/data
environment:
# PROXY_SERVER: "" # Proxy server address
# OPENAI_LOGIN_TYPE: "" # Login Type,nologin or email or google
# OPENAI_LOGIN_EMAIL: "" # Login email
# OPENAI_LOGIN_PASSWORD: "" # Login password
restart: unless-stoppedAll environment variables are optional.
| Variable | Description | Default |
|---|---|---|
| PROXY_SERVER | Proxy server address | None |
| DATA_DIR | Data storage directory | ./data |
| OPENAI_LOGIN_TYPE | ChatGPT login type: nologin, email, google |
nologin |
| OPENAI_LOGIN_EMAIL | Email account for email login | None |
| OPENAI_LOGIN_PASSWORD | Password for email login | None |
| OPENAI_LOGIN_APP_PASSWORD | App password for email verification | None |
| GOOGLE_LOGIN_EMAIL | Google login email | None |
| GOOGLE_LOGIN_PASSWORD | Google login password | None |
| GOOGLE_LOGIN_OTP_SECRET | Google login 2FA secret | None |
| GOOGLE_LOGIN_RECOVERY_EMAIL | Google login recovery email | None |
The system uses Playwright to control a fingerprint browser, simulating user operations to send requests to the OpenAI website and converting the responses into API interfaces.
Currently supports the OpenAI-compatible /v1/chat/completions API, which can be accessed using OpenAI or other compatible clients.
Chat completion API，compatible with OpenAI Chat Completions API。
POST /v1/chat/completions
Request：
{
"model": "gpt-4o",
"messages": [
{
"role": "user",
"content": "Hello"
}
],
// Optional: If using SSE stream, set to true, default is false
"stream": false
// Optional: If enabled, the response will include meta information(message_id and conversation id)
// If both parent_message_id and conversation_id are provided, the request will continue within the existing conversation context.
// If they are not set, the request will be treated as a new conversation.
// "meta": {
// "enable": true,
// "parent_message_id": "5363437e-b364-4b72-b3d6-415deeed11ab", # Optional
// "conversation_id": "6774f183-f70c-800b-9965-6c110d3a3485" # Optional
// }
}Response：
{
"id": "chatcmpl-fZc6l869OzRu8rp7X8Dhj0COfTsR6",
"object": "chat.completion",
"created": 1733726226,
"model": "gpt-4o",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Hi there! How can I assist you today? 😊"
},
"logprobs": null,
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 1,
"completion_tokens": 11,
"total_tokens": 12
},
"meta": {
"message_id": "dffd63ef-63ac-4d40-b6de-e33ec40de9e2",
"conversation_id": "6774f183-f70c-800b-9965-6c110d3a3485"
}
}Audio Speech API, compatible with OpenAI's Audio Speech API. It allows users to generate speech from text input.
POST /v1/audio/speech
Request:
{
"input": "Hello, how are you?",
"voice": "cove"
"model": "tts-1",
// Optional: specify response format, default is "aac"
"response_format": "mp3",
}Response:
If successful, the API returns an audio file in the requested format.
import openai
openai.api_key = 'anything'
openai.base_url = "http://localhost:5000/v1/"
completion = openai.chat.completions.create(
model="gpt-4o-mini",
messages=[
{"role": "user", "content": "Hello"},
],
)
print(completion.choices[0].message.content)import OpenAI from 'openai';
const openai = new OpenAI({
apiKey: "anything",
baseURL: "http://localhost:5000/v1/",
});
const chatCompletion = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Echo Hello' }],
model: 'gpt-4o-mini',
});
console.log(chatCompletion.choices[0].message.content);- chat2api: https://github.com/lanqian528/chat2api