Obsidian+AI new gameplay: free multi-round dialogue and model freedom

Written by
Silas Grey
Updated on:June-27th-2025
Recommendation

Obsidian+AI new gameplay, unlocking multi-round dialogue and model freedom, and efficiently integrating ChatGPT.

Core content:
1. The chatgpt-md plug-in integrates ChatGPT in Obsidian, supporting OpenAI, OpenRouter and Ollama models
2. Plug-in configuration and feature demonstration to achieve multi-round dialogue and content modification
3. Free AI service configuration and interface sharing to achieve efficient AI integration

Yang Fangxian
Founder of 53A/Most Valuable Expert of Tencent Cloud (TVP)

 

Obsidian+AI new gameplay: free multi-round dialogue and model freedom

If you want to talk to ChatGPT directly in Obsidian and change the content of the conversation, this chatgpt-md plugin can do it. Not only does it support OpenAI, but it can also connect to OpenRouter and local Ollama models. It is flexible and easy to use!

Plugin Introduction

  • • Plugin ID:  chatgpt-md
  • • Project address: https://github.com/bramses/chatgpt-md
  • • Project Description: Make ChatGPT and Obsidian integrate (almost) seamlessly.
  • • Support for configuring 3 types of models: Each AI service now has its own dedicated URL parameter in settings and metadata
    • • openaiUrl is used for OpenAI API
    • • openrouterUrl is used for OpenRouter.ai
    • • ollamaUrl is used for Ollama Local Model

chatgpt-md plugin demo

Last time, fans left a message saying that they didn’t know how to use this plug-in. This issue specially recorded the screen to give you a step-by-step tutorial, explaining each parameter clearly to you.

Demo content

  1. 1. chatgpt-md plugin configuration
  2. 2. Features of chatgpt-md plugin
    1. 1. Multi-round dialogueYou can see the content of each conversation in markdown and modify the content of the conversation at any time , which is an advantage that other plug-ins cannot achieve
    2. 2. All prompt words and model parameters are configured in one stop in the notebook , and can be executed after editing infer title Generate a title,Very compact experience;

Have you used the following AI plug-ins?


AI service configuration

Use free_chat_gpt_api and openrouter.ai to use AI for free. If you don't have enough credit, you can recharge and pay in open_router.

Free OpenAI API

Project: https://github.com/popjane/free_chatgpt_api

Use the following tick models completely free of charge:

  • • gpt-4o-mini (average speed, if you want to experience extremely fast response, you can purchase a paid API)
  • • gpt-3.5-turbo-0125
  • • gpt-3.5-turbo-1106
  • • gpt-3.5-turbo
  • • gpt-3.5-turbo-16k
  • • net-gpt-3.5-turbo (network search model - slightly less stable)
  • • whisper-1
  • • dall-e-2

Click to get a free API key: https://free.v36.cm/github API URL: https://free.v36.cm

OpenRouter's free aggregation API

Service: https://openrouter.ai

Openrouter provides the following free models for use

Google

Model Name
Context Window
Gemini 2.5 Pro Experimental
1,000,000
Gemini 2.0 Flash Experimental
1,048,576
Gemini 2.0 Flash Thinking Experimental
40,000
LearnLM 1.5 Pro Experimental
40,960
Gemma 3 27B / 12B / 4B / 1B
131,072 each (96,000 or 32,768 for some)
Gemma 2 9B
8,192

DeepSeek

Model Name
Context Window
DeepSeek V3 0324 / Base
163,840
DeepSeek Chat (Old Version)
163,840
R1 / R1 Zero
163,840
R1 Distill Llama 70B
128,000
R1 Distill Qwen 32B / 14B
16,000 / 64,000

NVIDIA

Model Name
Context Window
Llama 3.1 Nemotron Ultra 253B
131,072
Llama 3.1 Nemotron 70B Instruct
131,072
Llama 3.1 Nemotron Nano 8B
131,072
Llama 3.3 Nemotron Super 49B
131,072

Meta (Llama series)

Model Name
Context Window
Llama 3.3 70B Instruct
8,000
Llama 3.2 Vision 11B
131,072
Llama 3.2 3B / 1B
20,000 / 131,072
Llama 3.1 405B (base)
64,000
Llama 3.1 8B Instruct
131,072
Llama 4 Maverick
256,000
Llama 4 Scout
512,000

Qwen (Ali)

Model Name
Context Window
Qwen2.5 7B / 72B / 32B Coder
32,768
Qwen2.5 VL 72B/32B/7B/3B
131,072 / 8,192 / 64,000 / 64,000
QwQ 32B / Preview
40,000 / 16,384

Conclusion

Whether it is modifying conversation records in real time, switching models at any time, or using the free API to take advantage of GPT-4o,chatgpt-md The design of the plug-in is precisely to meet the needs of knowledge workers. You can chat with AI while taking notes without switching to the website. It is free and saves you time and money. I recommend you to try it.