Discord Bot

Build a Discord bot with discord.py and the OpenAI Python SDK using AICredits as the OpenAI-compatible API endpoint.

Use discord.py with the OpenAI Python SDK to build a Discord bot that calls AICredits. Each message is billed through your AICredits wallet and appears in the usage dashboard.

Setup

Install the libraries in your bot project:

pip install discord.py openai

Set secrets as environment variables:

export DISCORD_BOT_TOKEN="your-discord-bot-token"
export AICREDITS_API_KEY="sk-your-aicredits-key"

Bot Example

bot.py

import asyncio
import os
from collections import defaultdict, deque

import discord
from openai import AsyncOpenAI, APIStatusError

client = AsyncOpenAI(
    base_url="https://api.aicredits.in/v1",
    api_key=os.environ["AICREDITS_API_KEY"],
)

intents = discord.Intents.default()
intents.message_content = True
bot = discord.Client(intents=intents)

history_by_channel: dict[int, deque[dict[str, str]]] = defaultdict(lambda: deque(maxlen=12))


@bot.event
async def on_ready():
    print(f"Logged in as {bot.user}")


@bot.event
async def on_message(message: discord.Message):
    if message.author.bot:
        return

    if bot.user not in message.mentions:
        return

    channel_history = history_by_channel[message.channel.id]
    prompt = message.content.replace(f"<@{bot.user.id}>", "").strip()
    channel_history.append({"role": "user", "content": prompt})

    messages = [
        {"role": "system", "content": "You are a concise Discord assistant."},
        *channel_history,
    ]

    async with message.channel.typing():
        try:
            response = await client.chat.completions.create(
                model="openai/gpt-4o-mini",
                messages=messages,
                max_tokens=600,
            )
        except APIStatusError as exc:
            if exc.status_code == 402:
                await message.reply("AICredits wallet balance is low or this key hit its budget limit.")
                return
            if exc.status_code == 429:
                await message.reply("Rate limit reached. Please try again in a moment.")
                return
            raise

    answer = response.choices[0].message.content or ""
    channel_history.append({"role": "assistant", "content": answer})
    await message.reply(answer[:1900])


asyncio.run(bot.start(os.environ["DISCORD_BOT_TOKEN"]))

Token and Cost Controls

Set max_tokens on every request so one Discord message cannot produce an unexpectedly long reply.
Keep a short per-channel history, such as the last 6 user/assistant turns.
Use a lower-cost default model like openai/gpt-4o-mini, then switch specific commands to larger models when needed.
Set an API-key budget cap in Dashboard → API Keys for bot deployments.

Verify

Mention the bot in a channel:

@your-bot Reply with "AICredits via Discord" and nothing else.

Then check Dashboard → Usage for the model, tokens, latency, and INR cost.

Troubleshooting

402 Payment Required — Add wallet credits or raise the key's budget cap.

429 Too Many Requests — Reduce parallel bot requests, add a queue, or lower the bot's Discord command fan-out.

Bot never responds — Enable the Discord message content intent for your bot and confirm the bot is invited with permission to read and send messages.

Setup

Bot Example

Token and Cost Controls

Verify

Troubleshooting

On this page