Repository navigation
[Feedback]:聽#37
Description
馃З Feedback
Hello BizzlZ Team,
I hope you are doing well. First of all, I truly appreciate the amazing work you are doing by providing such powerful AI services for free. It really helps developers and students like us a lot.
I wanted to kindly request one important feature that would make your platform even better. If possible, please consider adding OpenAI SDK compatible API support (like /v1/chat/completions format).
Today, most modern AI tools and agent frameworks (such as @openai/agents, LangChain, Vercel AI SDK, etc.) are built around the OpenAI-compatible API standard. Platforms like Groq and Google already support this, which makes integration extremely easy.
If BizzlZ also becomes compatible with the OpenAI SDK format, many developers will be able to use your models directly in their existing projects without changing their code. This would significantly increase adoption and usage of your platform.
Once again, thank you for your generosity and for supporting the developer community. I would really appreciate it if you could consider this request.
Activity
swaritsharma001 commentedon Dec 7, 2025
import { MultiModelAgent } from "multimodelagent";
import { tool } from "@openai/agents";
import { z } from "zod";
import axios from "axios";
const googleSearch = tool({
name: "google_search",
description: "Search anything using Serper API.",
parameters: z.object({
q: z.string(),
}),
async execute({ q }) {
const response = await axios.post(
"https://google.serper.dev/search",
{ q },
{
headers: {
"X-API-KEY": "YOUR_SERPER_API_KEY",
"Content-Type": "application/json",
},
}
);
return response.data;
},
});
const fetchChunk = tool({
name: "fetch_chunk",
description: "Fetch a small chunk from any webpage.",
parameters: z.object({
url: z.string(),
index: z.number(),
}),
async execute({ url, index }) {
return {
chunk:
Chunk ${index} from ${url},hasMore: index < 2,
};
},
});
const ai = new MultiModelAgent({
apiKey: "key",
model: "Qwen/Qwen3-4B",
baseURL: "https://api.bytez.com/models/v2/",
tools: [googleSearch, fetchChunk],
instructions: "You are a helpful AI assistant.",
onToolCall: (name, args) => {
console.log("TOOL CALL =>", name, args);
},
});
const reply = await ai.ask("who are you?");
console.log("AI:", reply)
inf3rnus commentedon Jan 6, 2026
@swaritsharma001 Hey! 馃憢
We do support the chat completions spec, please see
https://docs.bytez.com/http-reference/examples/openai-compliant/chatCompletionsExample