Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-11-25 17:29:22 +01:00
Code Issues Packages Projects Releases Wiki Activity

Manually install CUDA runtime libraries

Browse Source
This commit is contained in:
oobabooga 2023-10-12 21:02:44 -07:00
parent faf5c4dd58
commit 43be1be598
2 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
README.md
View File

@ -1,8 +1,3 @@
**Breaking change: WebUI now uses PyTorch 2.1.**
* For one-click installer users: If you encounter problems after updating, rerun the update script. If issues persist, delete the `installer_files` folder and use the start script to reinstall requirements.
* For manual installations, update PyTorch with the [provided command](https://github.com/oobabooga/text-generation-webui/#2-install-pytorch).
# Text generation web UI
A Gradio web UI for Large Language Models.
@ -100,6 +95,12 @@ conda activate textgen
The up-to-date commands can be found here: https://pytorch.org/get-started/locally/.
For NVIDIA, you may also need to manually install the CUDA runtime libraries:
```
conda install -y -c "nvidia/label/cuda-11.8.0" cuda-runtime
```
#### 3. Install the web UI
```

4
one_click.py
View File

@ -187,6 +187,10 @@ def install_webui():
# Install Git and then Pytorch
run_cmd(f"{install_git} && {install_pytorch} && python -m pip install py-cpuinfo==9.0.0", assert_success=True, environment=True)
# Install CUDA libraries (this wasn't necessary for Pytorch before...)
if choice == "A":
run_cmd("conda install -y -c \"nvidia/label/cuda-11.8.0\" cuda-runtime", assert_success=True, environment=True)
# Install the webui requirements
update_requirements(initial_installation=True)