Awesome
<h1 align="center"> llmcord </h1> <h3 align="center"><i> Talk to LLMs with your friends! </i></h3> <p align="center"> <img src="https://github.com/jakobdylanc/llmcord/assets/38699060/789d49fe-ef5c-470e-b60e-48ac03057443" alt=""> </p>llmcord turns Discord into a collaborative LLM frontend. It works with practically any LLM, remote or locally hosted.
Features
Reply-based chat system
Just @ the bot to start a conversation and reply to continue. Build conversations with reply chains!
You can do things like:
- Continue your own conversation or someone else's
- "Rewind" a conversation by simply replying to an older message
- @ the bot while replying to any message in your server to ask a question about it
Additionally:
- Back-to-back messages from the same user are automatically chained together. Just reply to the latest one and the bot will see all of them.
- You can seamlessly move any conversation into a thread. Just create a thread from any message and @ the bot inside to continue.
- When DMing the bot, conversations continue automatically (no reply required). To start a fresh conversation, just @ the bot. You can still reply to continue from anywhere.
Choose any LLM
llmcord supports remote models from:
Or run a local model with:
...Or use any other OpenAI compatible API server.
And more:
- Supports image attachments when using a vision model (like gpt-4o, claude-3, llava, etc.)
- Supports text file attachments (.txt, .py, .c, etc.)
- Customizable personality (aka system prompt)
- User identity aware (OpenAI API and xAI API only)
- Streamed responses (turns green when complete, automatically splits into separate messages when too long)
- Hot reloading config (you can change settings without restarting the bot)
- Displays helpful warnings when appropriate (like "⚠️ Only using last 25 messages" when the customizable message limit is exceeded)
- Caches message data in a size-managed (no memory leaks) and mutex-protected (no race conditions) global dictionary to maximize efficiency and minimize Discord API calls
- Fully asynchronous
- 1 Python file, ~200 lines of code
Instructions
-
Clone the repo:
git clone https://github.com/jakobdylanc/llmcord -
Create a copy of "config-example.yaml" named "config.yaml" and set it up:
Discord settings:
| Setting | Description |
|---|---|
| bot_token | Create a new Discord bot at discord.com/developers/applications and generate a token under the "Bot" tab. Also enable "MESSAGE CONTENT INTENT". |
| client_id | Found under the "OAuth2" tab of the Discord bot you just made. |
| status_message | Set a custom message that displays on the bot's Discord profile. Max 128 characters. |
| allow_dms | Set to false to disable direct message access.<br />(Default: true) |
| allowed_channel_ids | A list of Discord channel IDs where the bot can be used. Also accepts category IDs. Leave empty to allow all channels. Does not affect DMs. |
| allowed_role_ids | A list of Discord role IDs that can use the bot. Leave empty to allow everyone. DMs are force-disabled when at least one role is specified. |
| blocked_user_ids | A list of Discord user IDs that are blocked from using the bot. |
| max_text | The maximum amount of text allowed in a single message, including text from file attachments.<br />(Default: 100,000) |
| max_images | The maximum number of image attachments allowed in a single message. Only applicable when using a vision model.<br />(Default: 5) |
| max_messages | The maximum number of messages allowed in a reply chain. When exceeded, the oldest messages in the reply chain are dropped.<br />(Default: 25) |
| use_plain_responses | When set to true the bot will use plaintext responses instead of embeds. Plaintext responses have a shorter character limit so the bot's messages may split more often. Also disables streamed responses and warning messages.<br />(Default: false) |
LLM settings:
| Setting | Description |
|---|---|
| providers | Add the LLM providers you want to use, each with a base_url and optional api_key entry. Popular providers (openai, ollama, etc.) are already included. Only supports OpenAI compatible APIs. |
| model | Set to <provider name>/<model name>, e.g:<br /><br />-openai/gpt-4o<br />-ollama/llama3.3<br />-openrouter/anthropic/claude-3.5-sonnet |
| extra_api_parameters | Extra API parameters for your LLM. Add more entries as needed. Refer to your provider's documentation for supported API parameters.<br />(Default: max_tokens=4096, temperature=1.0) |
| system_prompt | Write anything you want to customize the bot's behavior! Leave blank for no system prompt. |
-
Run the bot:
No Docker:
python -m pip install -U -r requirements.txt python llmcord.pyWith Docker:
docker compose up
Notes
-
If you're having issues, try my suggestions here
-
Only models from OpenAI API and xAI API are "user identity aware" because only they support the "name" parameter in the message object. Hopefully more providers support this in the future.
-
PRs are welcome :)