github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10227] Make GPU vendor support optional during installation on Windows #53221

New Issue
Open
opened 2026-04-29 02:24:02 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-29 02:24:02 -05:00
Owner
Copy Link

Originally created by @DimmaDont on GitHub (Apr 11, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10227

This is the issue report for #10186.

Problem

Currently on Windows, the Ollama installer installs both CUDA and ROCm files by default. This is unnecessary for users using GPUs from only a single vendor, and increases installation time and disk usage.

Importance

This will improve installation and update times for users, and decrease disk usage.
In testing, ROCm-only installs took only 33% of the time a full install takes, and CUDA-only 75%.
For disk space, ROCm-only users will save ~3GB and CUDA-only ~1.8GB.

Usage

At installation time, users can deselect unwanted libraries. For example, NVIDIA-only users can skip installing ROCm files and AMD-only users CUDA files. Users running both can select both, and users running CPU-only can deselect both.

Testing

Testing can be performed by compiling and running the Inno Setup file in the pull request.

When installing, check:

  • Selection of neither/one/other/both GPU vendor options.
  • Only the selected vendors' support files are installed.

Documentation

docs/windows.md could be updated to document the Inno Setup /TYPE= option that can now be used.

Originally created by @DimmaDont on GitHub (Apr 11, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10227 This is the issue report for #10186. ## Problem Currently on Windows, the Ollama installer installs both CUDA and ROCm files by default. This is unnecessary for users using GPUs from only a single vendor, and increases installation time and disk usage. ## Importance This will improve installation and update times for users, and decrease disk usage. In testing, ROCm-only installs took only **33%** of the time a full install takes, and CUDA-only 75%. For disk space, ROCm-only users will save ~3GB and CUDA-only ~1.8GB. ## Usage At installation time, users can deselect unwanted libraries. For example, NVIDIA-only users can skip installing ROCm files and AMD-only users CUDA files. Users running both can select both, and users running CPU-only can deselect both. ## Testing Testing can be performed by compiling and running the Inno Setup file in the pull request. When installing, check: * Selection of neither/one/other/both GPU vendor options. * Only the selected vendors' support files are installed. ## Documentation `docs/windows.md` could be updated to document the Inno Setup `/TYPE=` option that can now be used.
GiteaMirror added the feature request label 2026-04-29 02:24:02 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#53221
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?