github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #9184] AMD iGPU Radeon 680M not being used #5983

New Issue
Open
opened 2026-04-12 17:20:06 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 17:20:06 -05:00
Owner
Copy Link

Originally created by @ENTPRESTIGIOUS on GitHub (Feb 18, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9184

What is the issue?

From Reddit , I've noticed that people were able to get the Radeon 680M iGPU working on Ollama but not in Ubuntu. I added the HSA parameter but it seems to load slower than it runs on the CPU. I've noticed that a person using a 6700M which is a graphics card but older than my iGPU works https://github.com/ollama/ollama/issues/6152. I am not sure what to do and how to fix this because people have been saying that it doubles the speed which isn't the case for me. I have restarted the service and used daemon-reload and it doesn't seem to help.

In /etc/systemd/system/ollama.service

[Unit]
Description=Ollama Service
After=network-online.target

[Service]
ExecStart=/usr/local/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3
Environment="PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/snap/bin"
Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0"

[Install]
WantedBy=default.target

In /etc/systemd/system/ollama.service.d/override.conf

### Editing /etc/systemd/system/ollama.service.d/override.conf
### Anything between here and the comment below will become the contents of the drop-in file

[Service]
Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0"
Environment="OLLAMA_HOST=0.0.0.0"
Environment="OLLAMA_ORIGINS=*"

### Edits below this comment will be discarded


### /etc/systemd/system/ollama.service
# [Unit]
# Description=Ollama Service
# After=network-online.target
# 
# [Service]
# ExecStart=/usr/local/bin/ollama serve
# User=ollama
# Group=ollama
# Restart=always
# RestartSec=3
# Environment="PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/snap/bin"
# Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0"
# 
# [Install]

Relevant log output

Feb 17 18:31:50 willybee-EQ systemd[1]: Stopping ollama.service - Ollama Service...
Feb 17 18:31:50 willybee-EQ systemd[1]: ollama.service: Deactivated successfully.
Feb 17 18:31:50 willybee-EQ systemd[1]: Stopped ollama.service - Ollama Service.
Feb 17 18:31:50 willybee-EQ systemd[1]: ollama.service: Consumed 6min 767ms CPU time, 2.5G memory peak, 679.1M memory swap peak.
Feb 17 18:31:50 willybee-EQ systemd[1]: Started ollama.service - Ollama Service.
Feb 17 18:31:50 willybee-EQ ollama[460790]: 2025/02/17 18:31:50 routes.go:1186: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION:10.3.0 HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/usr/share/ollama/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[* http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.967-08:00 level=INFO source=images.go:432 msg="total blobs: 57"
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.968-08:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0"
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.968-08:00 level=INFO source=routes.go:1237 msg="Listening on [::]:11434 (version 0.5.11)"
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.969-08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.979-08:00 level=WARN source=amd_linux.go:61 msg="ollama recommends running the https://www.amd.com/en/support/linux-drivers" error="amdgpu version file missing: /sys/module/amdgpu/version stat /sys/module/amdgpu/version: no such file or directory"
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.980-08:00 level=INFO source=amd_linux.go:389 msg="skipping rocm gfx compatibility check" HSA_OVERRIDE_GFX_VERSION=10.3.0
Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.980-08:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=rocm variant="" compute=gfx1035 driver=0.0 name=1002:1681 total="4.0 GiB" available="4.0 GiB"
Feb 17 18:34:22 willybee-EQ ollama[460790]: [GIN] 2025/02/17 - 18:34:22 | 200 |     621.361µs |       127.0.0.1 | GET      "/api/version"

OS

Linux

GPU

AMD, Other

CPU

AMD

Ollama version

0.5.11

Originally created by @ENTPRESTIGIOUS on GitHub (Feb 18, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9184 ### What is the issue? From [Reddit](https://www.reddit.com/r/ollama/comments/1gu3akj/so_i_just_finished_setting_up_ollama_with_rocm/) , I've noticed that people were able to get the Radeon 680M iGPU working on Ollama but not in Ubuntu. I added the HSA parameter but it seems to load slower than it runs on the CPU. I've noticed that a person using a 6700M which is a graphics card but older than my iGPU works https://github.com/ollama/ollama/issues/6152. I am not sure what to do and how to fix this because people have been saying that it doubles the speed which isn't the case for me. I have restarted the service and used daemon-reload and it doesn't seem to help. In `/etc/systemd/system/ollama.service` ``` [Unit] Description=Ollama Service After=network-online.target [Service] ExecStart=/usr/local/bin/ollama serve User=ollama Group=ollama Restart=always RestartSec=3 Environment="PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/snap/bin" Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0" [Install] WantedBy=default.target ``` In `/etc/systemd/system/ollama.service.d/override.conf` ``` ### Editing /etc/systemd/system/ollama.service.d/override.conf ### Anything between here and the comment below will become the contents of the drop-in file [Service] Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0" Environment="OLLAMA_HOST=0.0.0.0" Environment="OLLAMA_ORIGINS=*" ### Edits below this comment will be discarded ### /etc/systemd/system/ollama.service # [Unit] # Description=Ollama Service # After=network-online.target # # [Service] # ExecStart=/usr/local/bin/ollama serve # User=ollama # Group=ollama # Restart=always # RestartSec=3 # Environment="PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin:/snap/bin" # Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0" # # [Install] ``` ### Relevant log output ```shell Feb 17 18:31:50 willybee-EQ systemd[1]: Stopping ollama.service - Ollama Service... Feb 17 18:31:50 willybee-EQ systemd[1]: ollama.service: Deactivated successfully. Feb 17 18:31:50 willybee-EQ systemd[1]: Stopped ollama.service - Ollama Service. Feb 17 18:31:50 willybee-EQ systemd[1]: ollama.service: Consumed 6min 767ms CPU time, 2.5G memory peak, 679.1M memory swap peak. Feb 17 18:31:50 willybee-EQ systemd[1]: Started ollama.service - Ollama Service. Feb 17 18:31:50 willybee-EQ ollama[460790]: 2025/02/17 18:31:50 routes.go:1186: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION:10.3.0 HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/usr/share/ollama/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[* http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.967-08:00 level=INFO source=images.go:432 msg="total blobs: 57" Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.968-08:00 level=INFO source=images.go:439 msg="total unused blobs removed: 0" Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.968-08:00 level=INFO source=routes.go:1237 msg="Listening on [::]:11434 (version 0.5.11)" Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.969-08:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.979-08:00 level=WARN source=amd_linux.go:61 msg="ollama recommends running the https://www.amd.com/en/support/linux-drivers" error="amdgpu version file missing: /sys/module/amdgpu/version stat /sys/module/amdgpu/version: no such file or directory" Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.980-08:00 level=INFO source=amd_linux.go:389 msg="skipping rocm gfx compatibility check" HSA_OVERRIDE_GFX_VERSION=10.3.0 Feb 17 18:31:50 willybee-EQ ollama[460790]: time=2025-02-17T18:31:50.980-08:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=rocm variant="" compute=gfx1035 driver=0.0 name=1002:1681 total="4.0 GiB" available="4.0 GiB" Feb 17 18:34:22 willybee-EQ ollama[460790]: [GIN] 2025/02/17 - 18:34:22 | 200 | 621.361µs | 127.0.0.1 | GET "/api/version" ``` ### OS Linux ### GPU AMD, Other ### CPU AMD ### Ollama version 0.5.11
GiteaMirror added the bug label 2026-04-12 17:20:06 -05:00
GiteaMirror commented 2026-04-12 17:20:07 -05:00
Author
Owner
Copy Link

@ENTPRESTIGIOUS commented on GitHub (Feb 18, 2025):

What confuses me more is that Linux identifies this iGPU as a Radeon 680m when it's a Ryzen 5 6600H which comes with a Radeon 660m

 lspci | grep VGA
05:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Rembrandt [Radeon 680M] (rev c9)
<!-- gh-comment-id:2664748102 --> @ENTPRESTIGIOUS commented on GitHub (Feb 18, 2025): What confuses me more is that Linux identifies this iGPU as a Radeon 680m when it's a Ryzen 5 6600H which comes with a Radeon 660m ``` lspci | grep VGA 05:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Rembrandt [Radeon 680M] (rev c9) ```
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#5983
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?