github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-05 23:53:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4614] Cpu selected over GPU when running ollama service #2898

New Issue
Closed
opened 2026-04-12 13:15:12 -05:00 by GiteaMirror · 7 comments
GiteaMirror commented 2026-04-12 13:15:12 -05:00
Owner
Copy Link

Originally created by @Talleyrand-34 on GitHub (May 24, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4614

Originally assigned to: @dhiltgen on GitHub.

What is the issue?

The problems is that i cannot specify to use gpu over cpu. The instructions to get the gpu running are also not clear.
Is it that the gpu is not supported? If yes which options do i have?

ollama ps
NAME                    ID              SIZE    PROCESSOR       UNTIL
codellama:latest        8fdf8f752f6e    5.1 GB  100% CPU        4 minutes from now

systemctl service

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

[Service]
Enviroment="OLLAMA_MODELS=/home/deck/.ollama/models"
ExecStart=/usr/local/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3
Environment="PATH=/root/.nix-profile/bin:/nix/var/nix/profiles/default/bin:/usr/local/sbin:/usr/local/bin:/usr/bin:/var/lib/flatpak/exports/bin:/usr/bin/site_perl:/usr/bin/vendor_perl:/usr/bin/core_perl:/usr/lib/rustup/bin"
Environment="OLLAMA_HOST=0.0.0.0"
Environment="OLLAMA_DEBUG=1"
#Enviroment="HSA_OVERRIDE_GFX_VERSION=gfx1030"

[Install]
WantedBy=default.target

systemctl status

May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_linux.go:243 msg="amdgpu memory" gpu=0 available="1.0 GiB"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /opt/rocm/lib"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/lib64"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/local/bin/rocm"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/share/ollama/lib/rocm"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=DEBUG source=amd_linux.go:292 msg="rocm supported GPUs" types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:296 msg="amdgpu is not supported" gpu=0 gpu_type=gfx1033 library=/usr/share/ollama/lib/rocm supported_types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:298 msg="See https://github.com/ollama/ollama/blob/main/docs/gpu.md#overrides for HSA_OVERRIDE_GFX_VERSION usage"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=amd_linux.go:311 msg="no compatible amdgpu devices detected"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=types.go:71 msg="inference compute" id=0 library=cpu compute="" driver=0.0 name="" total="14.5 GiB" available="2.8 GiB"

GPU info

04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405] (rev ae) (prog-if 00 [VGA controller])
        Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405]
        Flags: bus master, fast devsel, latency 0, IRQ 41
        Memory at f8e0000000 (64-bit, prefetchable) [size=256M]
        Memory at f8f0000000 (64-bit, prefetchable) [size=2M]
        I/O ports at 1000 [size=256]
        Memory at 80300000 (32-bit, non-prefetchable) [size=512K]
        Capabilities: [48] Vendor Specific Information: Len=08 <?>
        Capabilities: [50] Power Management version 3
        Capabilities: [64] Express Legacy Endpoint, MSI 00
        Capabilities: [a0] MSI: Enable- Count=1/4 Maskable- 64bit+
        Capabilities: [c0] MSI-X: Enable+ Count=4 Masked-
        Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
        Capabilities: [270] Secondary PCI Express
        Capabilities: [2b0] Address Translation Service (ATS)
        Capabilities: [2c0] Page Request Interface (PRI)
        Capabilities: [2d0] Process Address Space ID (PASID)
        Capabilities: [410] Physical Layer 16.0 GT/s <?>
        Capabilities: [440] Lane Margining at the Receiver <?>
        Kernel driver in use: amdgpu
        Kernel modules: amdgpu

I am using the steamdeck

OS

Linux // Arch linux

GPU

AMD

CPU

AMD

Ollama version

0.1.39

Originally created by @Talleyrand-34 on GitHub (May 24, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4614 Originally assigned to: @dhiltgen on GitHub. ### What is the issue? The problems is that i cannot specify to use gpu over cpu. The instructions to get the gpu running are also not clear. Is it that the gpu is not supported? If yes which options do i have? ``` ollama ps NAME ID SIZE PROCESSOR UNTIL codellama:latest 8fdf8f752f6e 5.1 GB 100% CPU 4 minutes from now ``` systemctl service ``` [Unit] Description=Ollama Service After=network-online.target [Service] Enviroment="OLLAMA_MODELS=/home/deck/.ollama/models" ExecStart=/usr/local/bin/ollama serve User=ollama Group=ollama Restart=always RestartSec=3 Environment="PATH=/root/.nix-profile/bin:/nix/var/nix/profiles/default/bin:/usr/local/sbin:/usr/local/bin:/usr/bin:/var/lib/flatpak/exports/bin:/usr/bin/site_perl:/usr/bin/vendor_perl:/usr/bin/core_perl:/usr/lib/rustup/bin" Environment="OLLAMA_HOST=0.0.0.0" Environment="OLLAMA_DEBUG=1" #Enviroment="HSA_OVERRIDE_GFX_VERSION=gfx1030" [Install] WantedBy=default.target ``` systemctl status ``` May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_linux.go:243 msg="amdgpu memory" gpu=0 available="1.0 GiB" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /opt/rocm/lib" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/lib64" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/local/bin/rocm" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/share/ollama/lib/rocm" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=DEBUG source=amd_linux.go:292 msg="rocm supported GPUs" types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:296 msg="amdgpu is not supported" gpu=0 gpu_type=gfx1033 library=/usr/share/ollama/lib/rocm supported_types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:298 msg="See https://github.com/ollama/ollama/blob/main/docs/gpu.md#overrides for HSA_OVERRIDE_GFX_VERSION usage" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=amd_linux.go:311 msg="no compatible amdgpu devices detected" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=types.go:71 msg="inference compute" id=0 library=cpu compute="" driver=0.0 name="" total="14.5 GiB" available="2.8 GiB" ``` GPU info ``` 04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405] (rev ae) (prog-if 00 [VGA controller]) Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405] Flags: bus master, fast devsel, latency 0, IRQ 41 Memory at f8e0000000 (64-bit, prefetchable) [size=256M] Memory at f8f0000000 (64-bit, prefetchable) [size=2M] I/O ports at 1000 [size=256] Memory at 80300000 (32-bit, non-prefetchable) [size=512K] Capabilities: [48] Vendor Specific Information: Len=08 <?> Capabilities: [50] Power Management version 3 Capabilities: [64] Express Legacy Endpoint, MSI 00 Capabilities: [a0] MSI: Enable- Count=1/4 Maskable- 64bit+ Capabilities: [c0] MSI-X: Enable+ Count=4 Masked- Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?> Capabilities: [270] Secondary PCI Express Capabilities: [2b0] Address Translation Service (ATS) Capabilities: [2c0] Page Request Interface (PRI) Capabilities: [2d0] Process Address Space ID (PASID) Capabilities: [410] Physical Layer 16.0 GT/s <?> Capabilities: [440] Lane Margining at the Receiver <?> Kernel driver in use: amdgpu Kernel modules: amdgpu ``` I am using the steamdeck ### OS Linux // Arch linux ### GPU AMD ### CPU AMD ### Ollama version 0.1.39
GiteaMirror added the gpubugamd labels 2026-04-12 13:15:12 -05:00
GiteaMirror commented 2026-04-12 13:15:14 -05:00
Author
Owner
Copy Link

@falmanna commented on GitHub (May 26, 2024):

Same here,

I am running wsl2 and was running fine, I upgraded my Nvidia driver yesterday and suddenly models are only using cpu. I don't see any errors in logs.

Ollama version: 0.1.38
Installation: docker
NVIDIA-SMI 555.42.03
Driver Version: 555.85
CUDA Version: 12.5

<!-- gh-comment-id:2132252103 --> @falmanna commented on GitHub (May 26, 2024): Same here, I am running wsl2 and was running fine, I upgraded my Nvidia driver yesterday and suddenly models are only using cpu. I don't see any errors in logs. Ollama version: 0.1.38 Installation: docker NVIDIA-SMI 555.42.03 Driver Version: 555.85 CUDA Version: 12.5
GiteaMirror commented 2026-04-12 13:15:14 -05:00
Author
Owner
Copy Link

@falmanna commented on GitHub (May 26, 2024):

UPDATE: Using llama.cpp directly is properly utilizing my gpu.

<!-- gh-comment-id:2132256344 --> @falmanna commented on GitHub (May 26, 2024): UPDATE: Using llama.cpp directly is properly utilizing my gpu.
GiteaMirror commented 2026-04-12 13:15:15 -05:00
Author
Owner
Copy Link

@Talleyrand-34 commented on GitHub (May 26, 2024):

Same here,

I am running wsl2 and was running fine, I upgraded my Nvidia driver yesterday and suddenly models are only using cpu. I don't see any errors in logs.

Ollama version: 0.1.38 Installation: docker NVIDIA-SMI 555.42.03 Driver Version: 555.85 CUDA Version: 12.5

I suggest activate debug and paste how ollama is choosing the methos as in here

May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_linux.go:243 msg="amdgpu memory" gpu=0 available="1.0 GiB"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /opt/rocm/lib"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/lib64"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/local/bin/rocm"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/share/ollama/lib/rocm"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=DEBUG source=amd_linux.go:292 msg="rocm supported GPUs" types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:296 msg="amdgpu is not supported" gpu=0 gpu_type=gfx1033 library=/usr/share/ollama/lib/rocm supported_types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:298 msg="See https://github.com/ollama/ollama/blob/main/docs/gpu.md#overrides for HSA_OVERRIDE_GFX_VERSION usage"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=amd_linux.go:311 msg="no compatible amdgpu devices detected"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=types.go:71 msg="inference compute" id=0 library=cpu compute="" driver=0.0 name="" total="14.5 GiB" available="2.8 GiB"
<!-- gh-comment-id:2132270361 --> @Talleyrand-34 commented on GitHub (May 26, 2024): > Same here, > > I am running wsl2 and was running fine, I upgraded my Nvidia driver yesterday and suddenly models are only using cpu. I don't see any errors in logs. > > Ollama version: 0.1.38 Installation: docker NVIDIA-SMI 555.42.03 Driver Version: 555.85 CUDA Version: 12.5 I suggest activate debug and paste how ollama is choosing the methos as in here ``` May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_linux.go:243 msg="amdgpu memory" gpu=0 available="1.0 GiB" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /opt/rocm/lib" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/lib64" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/local/bin/rocm" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/share/ollama/lib/rocm" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=DEBUG source=amd_linux.go:292 msg="rocm supported GPUs" types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:296 msg="amdgpu is not supported" gpu=0 gpu_type=gfx1033 library=/usr/share/ollama/lib/rocm supported_types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:298 msg="See https://github.com/ollama/ollama/blob/main/docs/gpu.md#overrides for HSA_OVERRIDE_GFX_VERSION usage" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=amd_linux.go:311 msg="no compatible amdgpu devices detected" May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=types.go:71 msg="inference compute" id=0 library=cpu compute="" driver=0.0 name="" total="14.5 GiB" available="2.8 GiB" ```
GiteaMirror commented 2026-04-12 13:15:16 -05:00
Author
Owner
Copy Link

@falmanna commented on GitHub (May 26, 2024):

Seems like the error is nvcuda init failure: 500

debug.log

<!-- gh-comment-id:2132379067 --> @falmanna commented on GitHub (May 26, 2024): Seems like the error is `nvcuda init failure: 500` [debug.log](https://github.com/ollama/ollama/files/15448749/debug.log)
GiteaMirror commented 2026-04-12 13:15:16 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (May 27, 2024):

@falmanna your issue is unrelated to this one. nVidia updated their drivers w/ version 555 and it's not loading the correct kernel drivers.

This is fixed in #4652 but the workaround for now is to run:

sudo modprobe nvidia
sudo modprobe nvidia_uvm

@Talleyrand-34 what is the model for your AMD card?

<!-- gh-comment-id:2132674594 --> @pdevine commented on GitHub (May 27, 2024): @falmanna your issue is unrelated to this one. nVidia updated their drivers w/ version 555 and it's not loading the correct kernel drivers. This is fixed in #4652 but the workaround for now is to run: ``` sudo modprobe nvidia sudo modprobe nvidia_uvm ``` @Talleyrand-34 what is the model for your AMD card?
GiteaMirror commented 2026-04-12 13:15:17 -05:00
Author
Owner
Copy Link

@Talleyrand-34 commented on GitHub (May 27, 2024):

It is the amd gpu of the steam deck, VanGogh.
The hardware complete info is the first message @pdevine

May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_linux.go:243 msg="amdgpu memory" gpu=0 available="1.0 GiB"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /opt/rocm/lib"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/lib64"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/local/bin/rocm"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/share/ollama/lib/rocm"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=DEBUG source=amd_linux.go:292 msg="rocm supported GPUs" types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:296 msg="amdgpu is not supported" gpu=0 gpu_type=gfx1033 library=/usr/share/ollama/lib/rocm supported_types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:298 msg="See https://github.com/ollama/ollama/blob/main/docs/gpu.md#overrides for HSA_OVERRIDE_GFX_VERSION usage"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=amd_linux.go:311 msg="no compatible amdgpu devices detected"
May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=types.go:71 msg="inference compute" id=0 library=cpu compute="" driver=0.0 name="" total="14.5 GiB" available="2.8 GiB"

GPU info

04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405] (rev ae) (prog-if 00 [VGA controller])
        Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405]
        Flags: bus master, fast devsel, latency 0, IRQ 41
        Memory at f8e0000000 (64-bit, prefetchable) [size=256M]
        Memory at f8f0000000 (64-bit, prefetchable) [size=2M]
        I/O ports at 1000 [size=256]
        Memory at 80300000 (32-bit, non-prefetchable) [size=512K]
        Capabilities: [48] Vendor Specific Information: Len=08 <?>
        Capabilities: [50] Power Management version 3
        Capabilities: [64] Express Legacy Endpoint, MSI 00
        Capabilities: [a0] MSI: Enable- Count=1/4 Maskable- 64bit+
        Capabilities: [c0] MSI-X: Enable+ Count=4 Masked-
        Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?>
        Capabilities: [270] Secondary PCI Express
        Capabilities: [2b0] Address Translation Service (ATS)
        Capabilities: [2c0] Page Request Interface (PRI)
        Capabilities: [2d0] Process Address Space ID (PASID)
        Capabilities: [410] Physical Layer 16.0 GT/s <?>
        Capabilities: [440] Lane Margining at the Receiver <?>
        Kernel driver in use: amdgpu
        Kernel modules: amdgpu
<!-- gh-comment-id:2133527259 --> @Talleyrand-34 commented on GitHub (May 27, 2024): It is the amd gpu of the steam deck, VanGogh. The hardware complete info is the first message @pdevine > ``` > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_linux.go:243 msg="amdgpu memory" gpu=0 available="1.0 GiB" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /opt/rocm/lib" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.970+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/lib64" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/local/bin/rocm" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.975+02:00 level=DEBUG source=amd_common.go:16 msg="evaluating potential rocm lib dir /usr/share/ollama/lib/rocm" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=DEBUG source=amd_linux.go:292 msg="rocm supported GPUs" types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:296 msg="amdgpu is not supported" gpu=0 gpu_type=gfx1033 library=/usr/share/ollama/lib/rocm supported_types="[gfx1030 gfx1100 gfx1101 gfx1102 gfx900 gfx906 gfx908 gfx90a gfx940 gfx941 gfx942]" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=WARN source=amd_linux.go:298 msg="See https://github.com/ollama/ollama/blob/main/docs/gpu.md#overrides for HSA_OVERRIDE_GFX_VERSION usage" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=amd_linux.go:311 msg="no compatible amdgpu devices detected" > May 24 20:10:14 steamdeck ollama[43437]: time=2024-05-24T20:10:14.977+02:00 level=INFO source=types.go:71 msg="inference compute" id=0 library=cpu compute="" driver=0.0 name="" total="14.5 GiB" available="2.8 GiB" > ``` > > GPU info > > ``` > 04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405] (rev ae) (prog-if 00 [VGA controller]) > Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] VanGogh [AMD Custom GPU 0405] > Flags: bus master, fast devsel, latency 0, IRQ 41 > Memory at f8e0000000 (64-bit, prefetchable) [size=256M] > Memory at f8f0000000 (64-bit, prefetchable) [size=2M] > I/O ports at 1000 [size=256] > Memory at 80300000 (32-bit, non-prefetchable) [size=512K] > Capabilities: [48] Vendor Specific Information: Len=08 <?> > Capabilities: [50] Power Management version 3 > Capabilities: [64] Express Legacy Endpoint, MSI 00 > Capabilities: [a0] MSI: Enable- Count=1/4 Maskable- 64bit+ > Capabilities: [c0] MSI-X: Enable+ Count=4 Masked- > Capabilities: [100] Vendor Specific Information: ID=0001 Rev=1 Len=010 <?> > Capabilities: [270] Secondary PCI Express > Capabilities: [2b0] Address Translation Service (ATS) > Capabilities: [2c0] Page Request Interface (PRI) > Capabilities: [2d0] Process Address Space ID (PASID) > Capabilities: [410] Physical Layer 16.0 GT/s <?> > Capabilities: [440] Lane Margining at the Receiver <?> > Kernel driver in use: amdgpu > Kernel modules: amdgpu > ``` >
GiteaMirror commented 2026-04-12 13:15:18 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (May 28, 2024):

Steam Deck (gfx1033) support is tracked via #3243

<!-- gh-comment-id:2135710625 --> @dhiltgen commented on GitHub (May 28, 2024): Steam Deck (gfx1033) support is tracked via #3243
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 amd bug gpu
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#2898
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?