github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15665] install.sh exits early on hybrid AMD + NVIDIA systems and skips AMD ROCm bundle #72052

New Issue
Open
opened 2026-05-05 03:24:10 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-05 03:24:10 -05:00
Owner
Copy Link

Originally created by @galuszkak on GitHub (Apr 18, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15665

What is the issue?

On machines like a Framework Laptop 16 with an NVIDIA dGPU and AMD Radeon 890M iGPU, the install flow is broken.

I'm using Fedora 43 - with both an NVIDIA GPU and an AMD iGPU/APU, scripts/install.sh exits as soon as nvidia-smi is detected and does not continue to the AMD detection path.

This means the script does not download and extract the ollama-linux--rocm package on hybrid systems, even when an AMD GPU is present.

From reading the current script the relevant logic is:

  • if nvidia-smi is available, print NVIDIA GPU installed. and exit 0
  • only after that, check for AMD and download ollama-linux-${ARCH}-rocm

Expected behavior:

  • On hybrid AMD + NVIDIA systems, the installer should not exit before checking AMD hardware.
  • If AMD hardware is detected, it should also install the Ollama ROCm bundle, or at minimum clearly document that hybrid systems require a separate manual ROCm bundle install.

Relevant log output

OS

Linux

GPU

AMD, Nvidia

CPU

AMD

Ollama version

0.21.0

Originally created by @galuszkak on GitHub (Apr 18, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15665 ### What is the issue? On machines like a Framework Laptop 16 with an NVIDIA dGPU and AMD Radeon 890M iGPU, the install flow is broken. I'm using Fedora 43 - with both an NVIDIA GPU and an AMD iGPU/APU, scripts/install.sh exits as soon as nvidia-smi is detected and does not continue to the AMD detection path. This means the script does not download and extract the ollama-linux-<arch>-rocm package on hybrid systems, even when an AMD GPU is present. From reading the current script the relevant logic is: - if nvidia-smi is available, print NVIDIA GPU installed. and exit 0 - only after that, check for AMD and download ollama-linux-${ARCH}-rocm Expected behavior: - On hybrid AMD + NVIDIA systems, the installer should not exit before checking AMD hardware. - If AMD hardware is detected, it should also install the Ollama ROCm bundle, or at minimum clearly document that hybrid systems require a separate manual ROCm bundle install. ### Relevant log output ```shell ``` ### OS Linux ### GPU AMD, Nvidia ### CPU AMD ### Ollama version 0.21.0
GiteaMirror added the bug label 2026-05-05 03:24:10 -05:00
GiteaMirror commented 2026-05-05 03:24:12 -05:00
Author
Owner
Copy Link

@PureBlissAK commented on GitHub (Apr 18, 2026):

🤖 Automated Triage & Analysis Report

Issue: #15665
Analyzed: 2026-04-18T18:13:39.119695

Analysis

  • Type: unknown
  • Severity: medium
  • Components: unknown

Implementation Plan

  • Effort: medium
  • Steps:

This issue has been triaged and marked for implementation.

<!-- gh-comment-id:4274294689 --> @PureBlissAK commented on GitHub (Apr 18, 2026): <!-- ollama-issue-orchestrator:v1 issue:15665 --> ## 🤖 Automated Triage & Analysis Report **Issue**: #15665 **Analyzed**: 2026-04-18T18:13:39.119695 ### Analysis - **Type**: unknown - **Severity**: medium - **Components**: unknown ### Implementation Plan - **Effort**: medium - **Steps**: *This issue has been triaged and marked for implementation.*
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 bug
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#72052
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?