This project is a simple script for interacting with the Chimera API. The main class is Chimera
, and it has methods for chat completion, image generation, text-to-speech generation, and content moderation.
- Node.js.
- Common Sense.
Install the project using npm:
npm install chimera-api
CommonJS Compatibility
This is an ESM module, you won't be able to use it with require()
. You will need to use import
instead:
(async () => {
const ChimeraModule = await import("chimera-api");
const Chimera = ChimeraModule.default;
const apiKey = "YOUR_API_KEY";
const chimera = new Chimera(apiKey);
const createChatCompletion = await chimera.chatCompletion({
model: "gpt-4",
messages: [
{
content: "Hello, how are you?",
role: "user",
},
],
});
console.log(createChatCompletion);
process.exit(0);
})();
First, import the Chimera
class:
import Chimera from "chimera-api";
Create a new instance of the class with your API key:
const chimera = new Chimera("your_api_key");
If you need to use a proxy, provide a configuration object as the second parameter to the constructor:
const chimera = new Chimera("your-api-key", {
host: "proxy-host",
port: "proxy-port",
protocol: "http",
auth: {
username: "proxy-username",
password: "proxy-password",
}, // optional
});
You can then use the class methods as follows:
let chatCompletionRequest = {
model: "gpt-4",
messages: [{ role: "user", content: "Hello" }],
// other properties...
};
chimera
.chatCompletion(chatCompletionRequest)
.then((response) => console.log(response))
.catch((error) => console.error(error));
Any other model that you want to use might follow this code aswell.
let imageRequest = {
prompt: "E G G.",
// other properties...
};
chimera
.imageGeneration(imageRequest)
.then((response) => console.log(response))
.catch((error) => console.error(error));
let ttsRequest = {
text: "Hello, world!",
};
chimera
.textToSpeech(ttsRequest)
.then((response) => console.log(response))
.catch((error) => console.error(error));
chimera
.Moderation("some potentially offensive text")
.then((response) => console.log(response))
.catch((error) => console.error(error));
Translate
chimera
.Whisper(
{
file: createReadStream("./path/to/file.mp3"),
fileName: "name.mp3",
language: "pt-BR",
},
"audio/translations"
)
Transcription
chimera
.Whisper(
{
file: createReadStream("./path/to/file.mp3"),
fileName: "name.mp3",
language: "pt-BR",
},
"audio/transcriptions"
)
There are also some examples on the test
folder.
Performance Related Stuff
The Proxy handler is 100% not a thing to be used with a single API: It provides advanced management of Axios instances, such as per-URL instance caching and optional proxy support. Originally, it was designed for managing multiple API endpoints, hence the proxy handler might seem over-engineered for a single API wrapper. It's optional and doesn't affect the main functionality if not utilized.
TLDR: The Proxy Handler is way too overengineered for a single API wrapper, but it's optional and doesn't affect the main functionality if not utilized.
Support
Mainly on the Discord Server You can also create an issue if what you're facing is an error from this package. (Please, do not ask for help on the discord if this is a package-related issue, this is not in any way attributed to the main devs on Chimera.)
This project is licensed under the GNU General Public License v3.0.
Please note that this is not an official project of the developers of Chimera. This was made out of personal interest and may be helpful to others. This comes with no guarantees or support, but feel free to use it as you wish within the bounds of the license.
Enjoy!
me made dis :)