10000 Ollama · Issue #37 · infiolab/infio-copilot · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Ollama #37

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wenrolland opened this issue Apr 11, 2025 · 4 comments
Open

Ollama #37

wenrolland opened this issue Apr 11, 2025 · 4 comments

Comments

@wenrolland
Copy link

Great plugin! I can't seem to get Ollama working.

Thanks.
@duanfuxiang0
Copy link
Member

May I ask which version you're using, 0.1.x? What error did you encounter?
The 0.1.x version works fine, but I only used it as an embedding model. As a chat model, the capabilities provided by ollama are still too weak.

@wenrolland
Copy link
Author

Hi, thanks for your answer. I'm using Ollama version is 0.6.5, the latest version. I just want to use it for completion and chat. I use many apps that work great with the available capabilities offered by Ollama. It's pretty amazing with the latest models like Gemma 3 and DeepSeek. In Obsidian, I'm using companion with great results for completion, but the app doesn't have a chat function and the completion context can't be easily adjusted to make it larger. I write stories and code, so I need to have maximum context for it to be coherent. :) Thanks again!

@duanfuxiang0
Copy link
Member

I'm using companion with great results for completion

Got it, I'll go learn about it.

I've also tested Gemma 3 and DeepSeek R1, but the results don't seem ideal. Could you share which specific model you're using? Is it 30b or 70b?
7D73

@wenrolland
Copy link
Author

I'm certainly not saying that I get the same results as Gemini, ChatGPT or Claude... :) My GPU can only handle up to 12b-13b with a usable speed. For programming, I'm using Yi-coder 9b for coding and for writing and document analysis : Gemma 3 12b and 4b for multimodal, DeepSeek R1 7b for thinking, Fireblossom, Aya-expanse and Lama 3.1, all 8b for writing and translating. My use is mainly around language, writing, coding and I use my own information for knowledge. For me, having Ollama as an option is great for privacy and when I'm on the move with no Internet connection. It's also great on the bank and when I want to write a story that include NSFW sections.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@duanfuxiang0 @wenrolland
0