[FIX]I have imported llama3.1, but it still cannot be used

[YJExpand]

Describe the bug

According to the document https://docs.khoj.dev/advanced/ollama, follow the steps to execute the terminal command, but it still doesn't work

To Reproduce

Steps to reproduce the behavior:

1. Download Ollama
2. ollama pull llama3.1
3. Uncomment OPENAI_API_BASE environment variable in your downloaded Khoj docker-compose.yml
4. docker-compose up
5. khoj: 🌖 Khoj is ready to use main.py:220
6. Enter text, but no reply

Screenshots

If applicable, add screenshots to help explain your problem.

Platform

• Server:
  • Cloud-Hosted (https://app.khoj.dev)
  • Self-Hosted Docker
  • Self-Hosted Python package
  • Self-Hosted source code
• Client:
  • Obsidian
  • Emacs
  • Desktop app
  • Web browser
  • WhatsApp
• OS:
  • Windows
  • macOS 14.1.1
  • CPU 3 GHz Hexa-core Intel Core i5
  • Radeon Pro 570X 4 GB
  • Linux
  • Android
  • iOS

If self-hosted

• Server Version [e.g. 1.0.1]: khoj 1.33.2

Additional context

Add any other context about the problem here.

[sabaimran]

Hi @YJExpand ! Thank you for all the details.

This isn't super clear in the docs, but you will have to add a different chat model that's pulling from Ollama, rather than the defaults. For instance, see the instructions for setting up with Ollama and in the Update tab, step 4.

Once you've configured that, use that to configure your model in your settings page.

[YJExpand]

Hi @YJExpand ! Thank you for all the details.

This isn't super clear in the docs, but you will have to add a different chat model that's pulling from Ollama, rather than the defaults. For instance, see the instructions for setting up with Ollama and in the Update tab, step 4.

Once you've configured that, use that to configure your model in your settings page.

Thank you very much. It is finally working, but it times out easily and blocks access to the server. Is it because of my device?

[sabaimran]

Okay glad we got the first issue working!

What kind of device do you have? Does it have a GPU, how much RAM?

There were some timeout-related limitations that we were going to expand that should help. We have fixed for those that should be in the next release.

[YJExpand]

Okay glad we got the first issue working!

What kind of device do you have? Does it have a GPU, how much RAM?

There were some timeout-related limitations that we were going to expand that should help. We have fixed for those that should be in the next release.

iMac
CPU: 3 GHz Hexa-core Intel Core i5
GPU: Radeon Pro 570X 4 GB
RAM: 32G