Glad it helped. I'm finding that leveraging AI in the Canvas is great. I can see the complete picture from start to finish, from my first question or idea to my final thoughts, all in one place.
I really liked the video, however in my case the valt indexing is taking a long time. The estimate is more than two weeks. Does this have to do with the size of the valt or the hardware?
@Antone, did you have any struggles with getting Llama to index your vault using Co-Pilot plugin? I followed all instructions but it gives me message that Co-pilot is indexing my vault but the count never goes above 0.
@@antoneheyward same issue you were having in the video. Using Copilot plugin with Ollama and Llama3. I can use the chat but not the Vault QA. It starts a reindex but never progresses. I also tried the Smart Second Brain and it does seem to progress with index but VERY slowly (telling me it will take over 300 hours!). I'm thinking my Macbook might not be powerful enough (it's M1 with 16 Gb).
The plugins i tested will not but you can ask it to create tags based on your note then add the tags tha it recommends. With that said, I think there is a plugin that will do this, not sure though.
It can be useful for some of the basic actions like correcting grammar, punctuation, getting facts about a topic, expanding on ideas and so on. So i don’t have to leave obsidian to use chatgpt or one of the other chatbot solutions. But using it to create a “smart second brain” seems to have too much friction right now.
![*Next-door 10x Software Engineer* [FULL]](/img/1.gif)
A good overview, thanks! - I’ve been wanting something like this for Obsidian, but haven’t had the time to sort it out myself.
Glad it helped. I'm finding that leveraging AI in the Canvas is great. I can see the complete picture from start to finish, from my first question or idea to my final thoughts, all in one place.
Really thorough testing. Thank you.
Glad you liked it! And appreciate the feedback.
this was great
Thanks for the positive feedback.
Thanks very much for the valuable compilation. Would you make the obsidian canvas from the presentation available for me / personal use?
I really liked the video, however in my case the valt indexing is taking a long time. The estimate is more than two weeks. Does this have to do with the size of the valt or the hardware?
Wow! I had a similar message when I first tried but changing the “embedding” model reduced the time estimate. Try changing it to see.
@@antoneheyward Thanks
I was just thinking about this ngl
Hope the video has useful information for you.
Small correction: It should be Cannoli instead of Canoli.
Thanks. Can-no-li instead of Ca-no-li.
这个流程框架图很棒
Thanks for the positive feedback.
@Antone, did you have any struggles with getting Llama to index your vault using Co-Pilot plugin? I followed all instructions but it gives me message that Co-pilot is indexing my vault but the count never goes above 0.
Which “embedding” model are you using? If I did have an issue it was related to that. But I don’t recall anything specific.
@@antoneheyward same issue you were having in the video. Using Copilot plugin with Ollama and Llama3. I can use the chat but not the Vault QA. It starts a reindex but never progresses. I also tried the Smart Second Brain and it does seem to progress with index but VERY slowly (telling me it will take over 300 hours!). I'm thinking my Macbook might not be powerful enough (it's M1 with 16 Gb).
Embedding model = "ollama-nomic-embed-text"
i have M1 Pro with 16gb. I have 1900 files in my vault. Maybe your vault as more? You can try different modals like mistral and mxbai-embed-large.
@@antoneheyward I have about the same size vault. I'll play around more with it, thanks for getting back to me.
Can it do stuff for you like puting tags on notes, stuff like that ?
The plugins i tested will not but you can ask it to create tags based on your note then add the tags tha it recommends. With that said, I think there is a plugin that will do this, not sure though.
Looks to me that chatgpt in obsidian is a solution looking for a problem
It can be useful for some of the basic actions like correcting grammar, punctuation, getting facts about a topic, expanding on ideas and so on. So i don’t have to leave obsidian to use chatgpt or one of the other chatbot solutions. But using it to create a “smart second brain” seems to have too much friction right now.