Pure Chat LLM leverages Obsidian notes to provide a graphical user interface (GUI) for ChatGPT directly within your vault. This plugin transforms your notes into interactive chat windows, enabling seamless conversations, brainstorming, questions, and prompt generation — all without leaving Obsidian.
- Interactive chat interface: Engage in seamless conversations with ChatGPT and other leading AI models directly from any Obsidian note. Edit your questions or responses and continue the chat without leaving your vault.
- Multiple AI providers: Choose from a wide range of supported model providers, including OpenAI, Gemini, xAI, Anthropic, Cohere, Mistral AI, and DeepSeek. Easily switch between providers and models to suit your needs.
- Note linking for context: Use
[[Note name]]or Markdown links to include the content of other notes in your chat, providing additional context or structuring complex prompts. (Referenced notes must start at the beginning of a line.) - Per-note customization: Configure API parameters (such as
model,max_completion_tokens, and more) on a per-note basis using a JSON block at the top of your note. - Flexible system prompts: Modify system prompts within your notes to influence the AI’s behavior for each conversation.
- Chat-linked notes: Link specific notes to individual chat sessions for organized, context-rich discussions.
- Markdown rendering: All chat responses support full Markdown formatting, including lists, code blocks, and more.
- Simple setup: Just add your API key for your chosen provider in the plugin settings to get started.
[!IMPORTANT] > Note: This plugin is not an AI note editor. It is designed solely for chatting with AI models through notes. It does not edit or manage your notes beyond the chat interface—it simply turns notes into GUI elements for AI chat.
- Create a new note in Obsidian.
- Write your question, prompt, or message.
- To initiate chat, press Shift + Enter or use the Complete chat response command.
- Your note will transform into a chat interface, allowing you to continue the conversation.
- To customize API parameters, add a JSON block at the top of your note, specifying options like:
model(e.g.,"gpt-4.1-nano")max_completion_tokensstream- Other API options (see OpenAI API documentation)
- To include content from other notes, use
[[Note name]]within your note. This will insert the entire content of that note at the position, which can be used as part of your message or role, helping you structure complex prompts or context.
Note
To include the content from another note, use the following JSON configuration:
{
"model": "gpt-4.1-nano",
"max_completion_tokens": 4096,
"stream": true
}Then, start your message with:
# role: system
You are a helpful assistant.
# role: user
[[Introduction]]
# role: assistant
Hello! How can I assist you today?(In this example, the content of the note named "Introduction" will be included in the message.)
Note: The note you reference must start at the beginning of a line in your markdown for this feature to work correctly.
- Open Settings -> Community Plugins
- Turn off Safe Mode
- Click Browse and search for Pure Chat LLM
- Click Install and then Enable
- (Optional) Configure plugin settings as desired
Before using the Pure Chat LLM plugin, you must enter your OpenAI API key:
- Navigate to Settings -> Community Plugins -> Pure Chat LLM
- Find the API key field
- Enter your OpenAI API key (you can generate one from the OpenAI API keys page)
- Save your settings
Note: The API key must be set before initiating chats. If you use the Complete chat response command without a configured API key, the plugin will prompt you to enter it.
- When you first try to run Complete chat response or start a new chat, if the API key isn't configured, you'll be prompted to enter it.
- After entering your API key, it will be saved for future sessions. You won't need to re-enter it unless you change it.