FreeAI::Server::OpenAI — OpenAI API server based on built-in AI model parsers.

Parser overview

The parser provides the ability to deploy your own OpenAI-compatible API server, which can be connected to from your applications (e.g., Cherry Studio, Cline, etc.) and scripts — both via the official OpenAI SDK and using regular HTTP requests. The parser provides access to free and paid models parsed by A-Parser.

List of supported models:

• FreeAI::ChatGPT
• FreeAI::Copilot
• FreeAI::DeepAI
• FreeAI::GoogleAI
• FreeAI::Kimi
• FreeAI::Perplexity

Go to DEMOBuy A-Parser Pro ($299)

Connection to Cherry Studio

• Settings (top right corner)
• List of providers, at the very bottom "add"
• Set an arbitrary name, Provider type must be OpenAI
• Enter API Key (any key)
• Enter host (configured in FreeAI::Server::OpenAI) initially http://127.0.0.1:3000
• "Manage" button to add the required models

View connection video

Connection via OpenAI SDK

Code from example

import OpenAI from "openai";

(async function () {
    const openai = new OpenAI({
        baseURL: "http://127.0.0.1:3000/v1", //Link where FreeAI::Server::OpenAI is hosted
        apiKey: "123", 
    });

    const completion = await openai.chat.completions.create({
        model: "FreeAI::ChatGPT", //Model is the parser name from the FreeAI::Server::OpenAI list, see "Parser overview" for supported models
        messages: [{ role: "user", content: "Why is the sky blue?" }], //Request to the model
    });

    console.log(completion.choices[0].message.content); //Output response from the AI model
})();

Getting results via HTTP request

Code from example

const resp = await fetch("http://127.0.0.1:3000/v1/chat/completions", {
    method: "POST",
    headers: {
        Authorization: "Bearer 123",
        "Content-Type": "application/json",
    },
    body: JSON.stringify({
        model: "FreeAI::ChatGPT", //Model is the parser name from the FreeAI::Server::OpenAI list, see "Parser overview" for supported models
        messages: [{ role: "user", content: "Nodejs use cases" }], //Request to the model
    }),
});

if (!resp.ok) {
    const text = await resp.text();
    throw new Error(`HTTP ${resp.status}: ${text}`);
}

const data = await resp.json();
console.log(data.choices?.[0]?.message?.content);

Possible settings

note

Common settings for all parsers

Parameter name	Default value	Description
Listen Host	127.0.0.1	IP address or hostname of the interface where the service accepts incoming connections
Listen Port	3000	Port number where the service accepts incoming connections
FreeAI::ChatGPT preset	default	Preset for FreeAI::ChatGPT parser
FreeAI::Copilot preset	default	Preset for FreeAI::Copilot parser
FreeAI::DeepAI preset	default	Preset for FreeAI::DeepAI parser
FreeAI::GoogleAI preset	default	Preset for FreeAI::GoogleAI parser
FreeAI::Kimi preset	default	Preset for FreeAI::Kimi parser
FreeAI::Perplexity preset	default	Preset for FreeAI::Perplexity parser