In this post, I’ll be sharing my experiences with the paid plan for the Obsidian Co-pilot plugin. This is based solely on my use of both the free and paid versions.
Obsidian Co-pilot is a community plugin, meaning it’s developed and maintained by the Obsidian community, not the core Obsidian developers. It adds Large Language Model capabilities to your Obsidian vault.
The ability to combine AI with your Personal Knowledge Management (PKM) system is intriguing and could lead to a truly operational personal Artificial Intelligence.
In this post we will discover if the paid version is worth it?
First impressions of Obsidian co-pilot and key features
I initially tried Obsidian Co-pilot while I was also using the Obsidian Smart Connections plugin, which has a similar aim of pairing your notes with AI. I preferred Co-pilot as it worked more smoothly for me. You can read the full story in my blog post Why I replaced Obsidian Smart Connections with Obsidian Co-pilot plugin.
The free version requires a licence key for the specific AI model you want to use, with support for many popular Large Language Models (LLMs)
As I already had an OpenAI licence key from using Smart Connections, I used it to connect Obsidian Co-pilot to an OpenAI model. I experienced no issues, and it worked well overall.
I had to pay OpenAI for the use of their model but as I wasn’t a heavy user at the time these costs were very reasonable.
I upgraded to the paid version to avoid the frustrating need to switch between querying my vault and general conversations with the AI, which was disrupting my workflow.
Another benefit is that the Co-pilot licence includes access to a dedicated Co-pilot model that manages the interaction with the external Large Language Model. This fixed cost allows me to predict my weekly expenses.
My experiences & Use cases with Obsidian Co-pilot
Before using Obsidian Co-pilot, you need to select the notes you want it to access.
I’ve given it access to notes in my Zettelkasten (permanent and literature notes), journals, my written content, tasks, projects, and ideas. It doesn’t access notes in my processing queue, as I only want it working with information I already know, even if I don’t immediately recall it.
Knowledge Management and my Zettelkasten
Obsidian houses my Personal Knowledge Management (PKM) system, with my Zettelkasten at its core. I don’t use Obsidian Co-pilot to process notes, as that process is about learning, reflecting on information, and distilling knowledge as I link ideas.
I’m concerned that using AI could negatively impact my learning and cognitive abilities, and I don’t want to risk that.
However, I do use it to ‘converse’ with my notes, mainly to identify knowledge gaps or areas where I already have some understanding.
For example, a few months ago, I wanted to improve my ability to ask Large Language Models the right questions. The output of this conversation helped the model better understand my knowledge gaps and create a more effective curriculum to address them.
I also use Obsidian co-pilot to summarise my notes and ideas.
Content creation
As I mentioned previously, I summarise my ideas and thoughts on a specific topic and then ask Obsidian Co-pilot to create a blog post layout based on that summary.
I created this blog post layout using this process, drawing on my existing notes and previous blog posts about the tool.
I then use this layout to write the first draft in Obsidian. The layout acts as a framework, but the creative process often leads the blog post to deviate from the original structure.
For example, the original layout had Knowledge Management and Zettelkasten after content creation, but I changed the order during writing to better reflect my workflow. AI is a valuable thinking partner, but the final decision is always yours.
After completing the first draft and letting it sit for a few days, I use Obsidian Co-pilot to edit and improve my blog post, asking it to explain the reasoning behind its suggested changes.
As someone who struggles with grammar and spelling, this is incredibly helpful – so much so that I won’t renew my grammar subscription when it expires.
If I’m unsure about which images to include, I ask Obsidian Co-pilot for suggestions. If I like an idea, I ask it to write a prompt for another AI to create the image.
Alternatives to Obsidian Co-pilot
In this section, I’ll compare Obsidian Co-pilot with Smart Connections, another AI plugin for Obsidian that I used previously. Like Obsidian Co-pilot, it’s a community plugin.
Obsidian Smart Connections offered similar core functionality to Obsidian Co-pilot, but I switched due to a combination of a better user interface and Smart Connections becoming unreliable. You can read the full story in my blog post “Why I switched from Obsidian Smart Connections to Obsidian Co-Pilot”.
If you use Obsidian, you might want to try both plugins to see which suits you best.
The next suggestion is a bit more unconventional, as it isn’t an Obsidian plugin. I probably wouldn’t have included it, but Obsidian Co-pilot picked up on my interest, likely due to a post I wrote about an experiment called “importing Obsidian files into Google Notebook LM”.
However, I’m currently testing Google Notebook LM as a way to collect notes on specific topics, as the 50-source limit isn’t an issue in this case. This could include material I’m researching, my own content, and notes from my Zettelkasten.
These notebooks supplement my Zettelkasten, helping me create podcasts and video overviews within Google NotebookLM to help improve my understanding of challenging subjects.
Obsidian Co-Pilot Pros and Cons
What I like about Obsidian Co-pilot:
- I can have a conversation with my notes
- Ability to identify notes that should be in the context of the conversation
- Allowing them to be the starting point for many of my blog posts
- It enhances my writing process especially around editing of the final post allowing me to cancel my Grammarly subscription
- Helps me to market my blog posts as effectively as I can with the short window I have available for that activity
- An easy to use UI
- It saves conversations
- Ability to update a note via a co-pilot conversation in canvas mode
Areas for improvement:
- Knowing which files are in context could be improved.
- The new canvas mode can be complex and has a steep learning curve.
Is Obsidian co-pilot worth paying for?
This is a difficult question to answer, as it depends on your specific needs. I recommend starting with the free version to see if it meets your requirements and whether the extra functionality of the paid version is worth the investment.
From my perspective, it’s definitely a worthwhile investment.
Further reading
My introductory guide to Zettelkasten: An introductory guide into the underlying Zettelkasten framework.
- Introductory guide to Large Language Model. An Introductory guide which explains what a Large Language model is
- Grammarly Review 2024. My most recent Grammarly review in which I share the benefits and the frustrations I found while using Grammarly.
- Why I replaced Obsidian Smart Connections with Obsidian Co-pilot plugin. In this blog post, I compare and co-pilot with smart connections and explain why I switched plugins.
- My experience of moving a large number of notes from Obsidian to Google Notebook LM. In this post, I explain my experiences of trying to move a large number of notes from Obsidian into Google Notebook LM