Nicholai/jan
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #1399 from janhq/gpuDocs

Browse Source 
docs: gpu not detected
This commit is contained in:
0xSage 2024-01-06 11:23:34 +08:00 committed by GitHub
commit 06ba70c661
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 49 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
docs/docs/guides/07-troubleshooting/02-failed-to-fetch.mdx
View File

@ -1,9 +1,8 @@
---
title: Failed To Fetch
title: Failed to Fetch
slug: /troubleshooting/failed-to-fetch
description: Troubleshooting "Failed to Fetch"
keywords:
[
keywords: [
Jan AI,
Jan,
ChatGPT alternative,
@ -17,19 +16,22 @@ keywords:
]
---
You may receive a "Error Occured: Failed to Fetch" response when you first start chatting with a selected model.
You may receive a "Error Occured: Failed to Fetch" response when you first start chatting with a selected model.
This may occur due to several reasons. Please follow these steps to resolve it:
1. Ensure you are on the latest version of Mac, Windows or Ubuntu OS version
- Upgrading to the latest version has resolved this issue for most people
2. Select a model that is smaller than 80% of your hardware V/RAM.
- For example, if you have a 8GB machine, you should select models smaller than 6GB.
- Upgrading to the latest version has resolved this issue for most people
2. Select a model that is smaller than 80% of your hardware V/RAM.
- For example, if you have a 8GB machine, you should select models smaller than 6GB.
3. Install the latest [Nightly release](https://jan.ai/install/nightly/)
- If you are re-installing Jan, it can help to [clear the application cache](https://jan.ai/troubleshooting/stuck-on-broken-build/)
4. Ensure your V/RAM is accessible by the application (some people have virtual RAM)
- If you are re-installing Jan, it can help to [clear the application cache](https://jan.ai/troubleshooting/stuck-on-broken-build/)
5. If you are on Nvidia GPUs, please download [Cuda](https://developer.nvidia.com/cuda-downloads)
4. Ensure your V/RAM is accessible by the application (some people have virtual RAM)
5. If you are on Nvidia GPUs, please download [Cuda](https://developer.nvidia.com/cuda-downloads)

37
docs/docs/guides/07-troubleshooting/03-gpu-not-used.mdx Normal file
View File

@ -0,0 +1,37 @@
---
title: Jan is Not Using GPU
keywords: [
Jan AI,
Jan,
ChatGPT alternative,
local AI,
private AI,
convZ
ersational AI,
no-subscription fee,
large language model,
troubleshooting,
]
---
By default, Jan uses GPU acceleration.
If you are experiencing this issue, you may need to download the latest Nvidia CUDA toolkit.
> We weren't propagating CUDA issues in the old build.
Steps:
1. Download Cuda toolkit (versions 11.4+) from [Nvidia](https://developer.nvidia.com/cuda-downloads?target_os=Windows)
1. Go to Settings > Advanced > Turn on GPU acceleration
1. You may need to restart Jan
1. Check in `System Monitor` that there is more (V)RAM now
1. If the issue persists, please install the [Nightly version](/install/nightly) instead.
1. If the issue persists, ensure your (V)RAM is accessible by the application. Some folks have virtual RAM and needs additional configuration.
1. Get help in [Jan Discord](https://discord.gg/mY69SZaMaC).