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 #8414] [Feature] Support Intel GPUs #67460

New Issue
Closed
opened 2026-05-04 10:26:32 -05:00 by GiteaMirror · 15 comments
GiteaMirror commented 2026-05-04 10:26:32 -05:00
Owner
Copy Link

Originally created by @NeoZhangJianyu on GitHub (Jan 14, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8414

Ollama had supported by the PR https://github.com/ollama/ollama/pull/2458 merged to support Intel GPU.
But the function disappears now.

I see there are several issues and opened PRs for Intel GPU. But they are too old.

I want to draft PRs to support Intel GPU: dGPU & iGPU (since 11th Core) by including llama.cpp SYCL backend.
This issue is created to trace the development work and reduce the duplicated work in the future.

Originally created by @NeoZhangJianyu on GitHub (Jan 14, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8414 Ollama had supported by the PR https://github.com/ollama/ollama/pull/2458 merged to support Intel GPU. But the function disappears now. I see there are several issues and opened PRs for Intel GPU. But they are too old. I want to draft PRs to support Intel GPU: dGPU & iGPU (since 11th Core) by including llama.cpp SYCL backend. This issue is created to trace the development work and reduce the duplicated work in the future.
GiteaMirror added the feature request label 2026-05-04 10:26:32 -05:00
GiteaMirror commented 2026-05-04 10:26:36 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (Jan 14, 2025):

@jmorganca @mchiang0610
Please assign this issue to me!

Thank you!

<!-- gh-comment-id:2588995455 --> @NeoZhangJianyu commented on GitHub (Jan 14, 2025): @jmorganca @mchiang0610 Please assign this issue to me! Thank you!
GiteaMirror commented 2026-05-04 10:26:38 -05:00
Author
Owner
Copy Link

@mchiang0610 commented on GitHub (Jan 14, 2025):

Hi @NeoZhangJianyu thank you for this. We are currently still early in our transition to Ollama's new engine. Ideally, you'd directly integrate with Ollama's new engine. This way, we don't have to continuously patch for backwards compatibility as we seek to improve Ollama's performance.

I just sent you a Slack invite to the channel we have with some Intel folks.

<!-- gh-comment-id:2589067831 --> @mchiang0610 commented on GitHub (Jan 14, 2025): Hi @NeoZhangJianyu thank you for this. We are currently still early in our transition to Ollama's new engine. Ideally, you'd directly integrate with Ollama's new engine. This way, we don't have to continuously patch for backwards compatibility as we seek to improve Ollama's performance. I just sent you a Slack invite to the channel we have with some Intel folks.
GiteaMirror commented 2026-05-04 10:26:39 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (Jan 14, 2025):

Yes, I have joined the Slack channel.

<!-- gh-comment-id:2589115723 --> @NeoZhangJianyu commented on GitHub (Jan 14, 2025): Yes, I have joined the Slack channel.
GiteaMirror commented 2026-05-04 10:26:41 -05:00
Author
Owner
Copy Link

@proline commented on GitHub (Feb 2, 2025):

I have the same problem, on one of my laptop I have an AMG integrated GPU that Ollama find, but on my other laptop where I have the intel processor ollama cannot use the integrated gpu.
When do you plan to add integated intel gpu to ollama ?

<!-- gh-comment-id:2629407779 --> @proline commented on GitHub (Feb 2, 2025): I have the same problem, on one of my laptop I have an AMG integrated GPU that Ollama find, but on my other laptop where I have the intel processor ollama cannot use the integrated gpu. When do you plan to add integated intel gpu to ollama ?
GiteaMirror commented 2026-05-04 10:26:42 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (Feb 5, 2025):

Support Intel iGPU is the scope of llama.cpp.
iGPU is limited by oneAPI support. It supports iGPU since 11th Core CPU.

Please refer to https://github.com/ggerganov/llama.cpp/blob/master/docs/backend/SYCL.md#hardware

Following iGPU is verified: iGPU in 13700k, i5-1250P, i7-1260P, i7-1165G7. But they are very slow. Please refer to: https://github.com/NeoZhangJianyu/ci_log/tree/main/llama.cpp

I suggest using dGPU or the built-in Arc GPU in Meteor Lake CPU or newer.

<!-- gh-comment-id:2635539278 --> @NeoZhangJianyu commented on GitHub (Feb 5, 2025): Support Intel iGPU is the scope of llama.cpp. iGPU is limited by oneAPI support. It supports iGPU since 11th Core CPU. Please refer to https://github.com/ggerganov/llama.cpp/blob/master/docs/backend/SYCL.md#hardware Following iGPU is verified: iGPU in 13700k, i5-1250P, i7-1260P, i7-1165G7. But they are very slow. Please refer to: https://github.com/NeoZhangJianyu/ci_log/tree/main/llama.cpp I suggest using dGPU or the built-in Arc GPU in Meteor Lake CPU or newer.
GiteaMirror commented 2026-05-04 10:26:44 -05:00
Author
Owner
Copy Link

@baoduy commented on GitHub (Feb 25, 2025):

Looking forward for this feature.

Thanks

<!-- gh-comment-id:2680912040 --> @baoduy commented on GitHub (Feb 25, 2025): Looking forward for this feature. Thanks
GiteaMirror commented 2026-05-04 10:26:45 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (May 25, 2025):

I have implemented to support Intel GPU on Linux.
Release for Windows is WIP.

I would like to merge to ollama official repo when it's totally ready and get approve by Ollama.

Refer to Intel GPU.
https://x.com/jianyu_neo/status/1926515682985775604

Thanks @zhewang1-intc for previous work to support Intel GPU.

<!-- gh-comment-id:2907630898 --> @NeoZhangJianyu commented on GitHub (May 25, 2025): I have implemented to support Intel GPU on Linux. Release for Windows is WIP. I would like to merge to ollama official repo when it's totally ready and get approve by Ollama. Refer to [Intel GPU](https://github.com/NeoZhangJianyu/ollama/blob/main/docs/intel_gpu.md). https://x.com/jianyu_neo/status/1926515682985775604 Thanks @zhewang1-intc for previous work to support Intel GPU.
GiteaMirror commented 2026-05-04 10:26:47 -05:00
Author
Owner
Copy Link

@AlanyTan commented on GitHub (Jun 23, 2025):

Is there any reason given why it's not yet merged to the official Ollama?

<!-- gh-comment-id:2996675080 --> @AlanyTan commented on GitHub (Jun 23, 2025): Is there any reason given why it's not yet merged to the official Ollama?
GiteaMirror commented 2026-05-04 10:26:48 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (Jun 24, 2025):

There is no plan to support Intel GPU in official Ollama.
Less Intel GPU user and need more devices to make sure the quality.

<!-- gh-comment-id:2998465583 --> @NeoZhangJianyu commented on GitHub (Jun 24, 2025): There is no plan to support Intel GPU in official Ollama. Less Intel GPU user and need more devices to make sure the quality.
GiteaMirror commented 2026-05-04 10:26:49 -05:00
Author
Owner
Copy Link

@technout commented on GitHub (Jul 6, 2025):

This blog about "Optimized Local AI on Intel Arc B580 with OpenWebUI and Ollama, using Fedora Linux and Podman"
It was very interesting, simple and it works very well on my own Linux Manjaro system with AMD Ryzen and Intel Arc A380 GPU
https://syslynx.net/llm-intel-b580-linux

<!-- gh-comment-id:3042287842 --> @technout commented on GitHub (Jul 6, 2025): This blog about "Optimized Local AI on Intel Arc B580 with OpenWebUI and Ollama, using Fedora Linux and Podman" It was very interesting, simple and it works very well on my own Linux Manjaro system with AMD Ryzen and Intel Arc A380 GPU https://syslynx.net/llm-intel-b580-linux
GiteaMirror commented 2026-05-04 10:26:52 -05:00
Author
Owner
Copy Link

@the-bort-the commented on GitHub (Jul 19, 2025):

So, this is basically dead in the water, @NeoZhangJianyu? Bad news as I was hoping to get support through apps on TrueNas, where I could use ollama when using Voice Assistant Preview Edition.

<!-- gh-comment-id:3092536156 --> @the-bort-the commented on GitHub (Jul 19, 2025): So, this is basically dead in the water, @NeoZhangJianyu? Bad news as I was hoping to get support through apps on TrueNas, where I could use ollama when using Voice Assistant Preview Edition.
GiteaMirror commented 2026-05-04 10:26:53 -05:00
Author
Owner
Copy Link

@yingding commented on GitHub (Jul 31, 2025):

I am looking forward to this too. Why is this issue closed?

<!-- gh-comment-id:3139644219 --> @yingding commented on GitHub (Jul 31, 2025): I am looking forward to this too. Why is this issue closed?
GiteaMirror commented 2026-05-04 10:26:55 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (Aug 4, 2025):

I provide a folk to support Intel GPU: Refer to Intel GPU.

https://x.com/jianyu_neo/status/1926515682985775604

<!-- gh-comment-id:3149563047 --> @NeoZhangJianyu commented on GitHub (Aug 4, 2025): I provide a folk to support Intel GPU: Refer to [Intel GPU](https://github.com/NeoZhangJianyu/ollama/blob/main/docs/intel_gpu.md). https://x.com/jianyu_neo/status/1926515682985775604
GiteaMirror commented 2026-05-04 10:26:58 -05:00
Author
Owner
Copy Link

@Chrisake commented on GitHub (Oct 9, 2025):

Can we still try to push the main repo to include the Intel GPUs even with errors at first? It seems that intel is currently offering the best vfm for AI at the market its a pity to have it already implemented and getting stuck into bureaucracy

<!-- gh-comment-id:3385168510 --> @Chrisake commented on GitHub (Oct 9, 2025): Can we still try to push the main repo to include the Intel GPUs even with errors at first? It seems that intel is currently offering the best vfm for AI at the market its a pity to have it already implemented and getting stuck into bureaucracy
GiteaMirror commented 2026-05-04 10:26:58 -05:00
Author
Owner
Copy Link

@NeoZhangJianyu commented on GitHub (Oct 10, 2025):

@Chrisake
You need to talk with ollama owners.

<!-- gh-comment-id:3387926045 --> @NeoZhangJianyu commented on GitHub (Oct 10, 2025): @Chrisake You need to talk with ollama owners.
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#67460
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?