Collection of free conversations about the big model inverse turn API interface

Free login GPT3.5 to API

https://github.com/missuo/FreeGPT35

https://github.com/aurora-develop/aurora

https://github.com/Dalufishe/freegptjs

https://github.com/PawanOsman/ChatGPT

https://github.com/nashsu/FreeAskInternet

https://github.com/aurorax-neo/free-gpt3.5-2api

https://github.com/aurora-develop/free-gpt3.5-2api

https://github.com/LanQian528/chat2api

https://github.com/k0baya/FreeGPT35-Glitch

https://github.com/cliouo/FreeGPT35-Vercel

https://github.com/hominsu/freegpt35

https://github.com/xsigoking/chatgpt-free-api

https://github.com/skzhengkai/free-chatgpt-api

https://github.com/aurora-develop/aurora-glitch (using glitch resources)

https://github.com/fatwang2/coze2openai (COZE to API, GPT4)

Domestic Model Reverse

search in depth(DeepSeek) Interface to API deepseek-free-api

Moonshot AI (Kimi.ai) Interface to API kimi-free-api

jumping stars (Leapfrog Ask StepChat) Interface to API step-free-api

Ali Tongyi (Qwen) Interface to API qwen-free-api

ZhipuAI (lit. record wisdom and say clearly) Interface to API glm-free-api

Meta AI (metaso) Interface to API metaso-free-api

Byte Jump (Beanbag) Interface to API doubao-free-api

Byte Jump (i.e. Dream AI) Interface to API jimeng-free-api

Spark Interface to API spark-free-api

MiniMax (Conch AI) Interface to API hailuo-free-api

Emohaa Interface to API emohaa-free-api

Login-free program with chat interface

https://github.com/Mylinde/FreeGPT

The work code for cloudflare, tie up your own domain name to play:

addEventListener("fetch", event => {
event.respondWith(handleRequest(event.request))
})

async function handleRequest(request) {
// Ensure that the request is a POST request and that the path is correct
if (request.method === "POST" && new URL(request.url).pathname === "/v1/chat/completions") {
const url = 'https://multillm.ai-pro.org/api/openai-completion'; // target API address
const headers = new Headers(request.headers);

// Add or modify headers as needed
headers.set('Content-Type', 'application/json');

// Get the body of the request and parse the JSON
const requestBody = await request.json();
const stream = requestBody.stream; // get the stream argument

// Construct a new request
const newRequest = new Request(url, {
method: 'POST',
headers: headers,
body: JSON.stringify(requestBody) // use modified body
});

try {
// Send a request to the target API
const response = await fetch(newRequest);

// Determine the response type based on the stream parameter
if (stream) {
// Handling streaming responses
const { readable, writable } = new TransformStream();
response.body.pipeTo(writable);
return new Response(readable, {
headers: response.headers
});
} else {
// Normal return response
return new Response(response.body, {
status: response.status,
headers: response.headers
});
}
} catch (e) {
// If the request fails, return an error message
return new Response(JSON.stringify({ error: 'Unable to reach the backend API' }), { status: 502 });
}
} else {
// Returns an error if the request method is not POST or the path is incorrect.
return new Response('Not found', { status: 404 });
}
}

POST Example:

curl --location 'https://ai-pro-free.aivvm.com/v1/chat/completions' \
--header 'Content-Type: application/json' \
--data '{
"model": "gpt-4-turbo".
"messages": [
{
"role": "user", "content": "Why Lu Xun Beat Zhou Shuren"
}],
"stream": true
}'

Add a pseudo-streaming code (output will be slower):

addEventListener("fetch", event => {
event.respondWith(handleRequest(event.request))
})

async function handleRequest(request) {
if (request.method === "OPTIONS") {
return new Response(null, {
headers: {
'Access-Control-Allow-Origin': '*',
"Access-Control-Allow-Headers": '*'
}, status: 204
})
}
// Ensure that the request is a POST request and that the path is correct
if (request.method === "POST" && new URL(request.url).pathname === "/v1/chat/completions") {
const url = 'https://multillm.ai-pro.org/api/openai-completion'; // target API address
const headers = new Headers(request.headers);

// Add or modify headers as needed
headers.set('Content-Type', 'application/json');

// Get the body of the request and parse the JSON
const requestBody = await request.json();
const stream = requestBody.stream; // get the stream argument

// Construct a new request
const newRequest = new Request(url, {
method: 'POST',
headers: headers,
body: JSON.stringify(requestBody) // use modified body
});

try {
// Send a request to the target API
const response = await fetch(newRequest);

// Determine the response type based on the stream parameter
if (stream) {
const originalJson = await response.json(); // read the full data at once
// Create a readable stream
const readableStream = new ReadableStream({
start(controller) {
// Send start data
const startData = createDataChunk(originalJson, "start");
controller.enqueue(new TextEncoder().encode('data: ' + JSON.stringify(startData) + '\n\n'));

// Assuming multiple chunks of data are processed and sent according to originalJson.
// For example, to simulate sending data in batches
const content = originalJson.choices[0].message.content; // assume this is the content to send
const newData = createDataChunk(originalJson, "data", content);
controller.enqueue(new TextEncoder().encode('data: ' + JSON.stringify(newData) + '\n\n'));

// Send end data
const endData = createDataChunk(originalJson, "end");
controller.enqueue(new TextEncoder().encode('data: ' + JSON.stringify(endData) + '\n\n'));

controller.enqueue(new TextEncoder().encode('data: [DONE]'));
// Mark the end of the stream
controller.close();
}
});
return new Response(readableStream, {
headers: {
'Access-Control-Allow-Origin': '*',
"Access-Control-Allow-Headers": '*',
'Content-Type': 'text/event-stream',
'Cache-Control': 'no-cache',
'Connection': 'keep-alive'
}
});
} else {
// Normal return response
return new Response(response.body, {
status: response.status,
headers: response.headers
});
}
} catch (e) {
// If the request fails, return an error message
return new Response(JSON.stringify({ error: 'Unable to reach the backend API' }), { status: 502 });
}
} else {
// Returns an error if the request method is not POST or the path is incorrect.
return new Response('Not found', { status: 404 });
}
}

// Create different data blocks based on type
function createDataChunk(json, type, content = {}) {
switch (type) {
case "start".
return {
id: json.id,
object: "chat.completion.chunk",
created: json.created,
model: json.model,
choices: [{ delta: {}, index: 0, finish_reason: null }]
};
case "data".
return {
id: json.id,
object: "chat.completion.chunk",
created: json.created,
model: json.model,
choices: [{ delta: { content }, index: 0, finish_reason: null }]
};
case "end".
return {
id: json.id,
object: "chat.completion.chunk",
created: json.created,
model: json.model,
choices: [{ delta: {}, index: 0, finish_reason: 'stop' }]
};
default.
return {};
}
}