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 #10506] Ollama not detecting AVX2 #6913

New Issue
Closed
opened 2026-04-12 18:48:12 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 18:48:12 -05:00
Owner
Copy Link

Originally created by @aaronoaks on GitHub (Apr 30, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10506

What is the issue?

I'm trying to run Ollama on a Dell Latitude laptop with integrated Intel Iris Xe graphics. I wasn't expecting any GPU support with that, but I found my testing was extremely slow. When I looked at the logs I saw it was using library=cpu variant="", not using AVX2 variant. I saw looking at other issues that most log outputs have a "Dynamic LLM libraries" line like this:

Nov 14 08:53:09 tal-ai-server ollama[1296]: time=2024-11-14T08:53:09.136+09:00 level=INFO source=common.go:49 msg="Dynamic LLM libraries" runners="[rocm cpu cpu_avx cpu_avx2 cuda_v11 cuda_v12]"

but a similar line does not appear in my log output. The CPU is an Intel Core i7-1365U, which Intel says supports AVX2.

Is there something I am missing to get Ollama to perform this detection and use AVX2?

Relevant log output

ollama serve
2025/04/30 12:09:09 routes.go:1232: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 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:C:\\Users\\aoaks\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE: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://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]"
time=2025-04-30T12:09:09.236-05:00 level=INFO source=images.go:458 msg="total blobs: 0"
time=2025-04-30T12:09:09.236-05:00 level=INFO source=images.go:465 msg="total unused blobs removed: 0"
time=2025-04-30T12:09:09.237-05:00 level=INFO source=routes.go:1299 msg="Listening on 127.0.0.1:11434 (version 0.6.6)"
time=2025-04-30T12:09:09.237-05:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
time=2025-04-30T12:09:09.237-05:00 level=INFO source=gpu_windows.go:167 msg=packages count=1
time=2025-04-30T12:09:09.237-05:00 level=INFO source=gpu_windows.go:183 msg="efficiency cores detected" maxEfficiencyClass=1
time=2025-04-30T12:09:09.238-05:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=10 efficiency=8 threads=12
time=2025-04-30T12:09:09.253-05:00 level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered"
time=2025-04-30T12:09:09.253-05:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="31.4 GiB" available="13.7 GiB"

OS

Windows

GPU

Intel

CPU

Intel

Ollama version

0.6.6

Originally created by @aaronoaks on GitHub (Apr 30, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10506 ### What is the issue? I'm trying to run Ollama on a Dell Latitude laptop with integrated Intel Iris Xe graphics. I wasn't expecting any GPU support with that, but I found my testing was extremely slow. When I looked at the logs I saw it was using `library=cpu variant=""`, not using AVX2 variant. I saw looking at other issues that most log outputs have a "Dynamic LLM libraries" line like this: > Nov 14 08:53:09 tal-ai-server ollama[1296]: time=2024-11-14T08:53:09.136+09:00 level=INFO source=common.go:49 msg="Dynamic LLM libraries" runners="[rocm cpu cpu_avx cpu_avx2 cuda_v11 cuda_v12]" but a similar line does not appear in my log output. The CPU is an Intel Core i7-1365U, which Intel says supports AVX2. Is there something I am missing to get Ollama to perform this detection and use AVX2? ### Relevant log output ```shell ollama serve 2025/04/30 12:09:09 routes.go:1232: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:2048 OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 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:C:\\Users\\aoaks\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE: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://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES:]" time=2025-04-30T12:09:09.236-05:00 level=INFO source=images.go:458 msg="total blobs: 0" time=2025-04-30T12:09:09.236-05:00 level=INFO source=images.go:465 msg="total unused blobs removed: 0" time=2025-04-30T12:09:09.237-05:00 level=INFO source=routes.go:1299 msg="Listening on 127.0.0.1:11434 (version 0.6.6)" time=2025-04-30T12:09:09.237-05:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" time=2025-04-30T12:09:09.237-05:00 level=INFO source=gpu_windows.go:167 msg=packages count=1 time=2025-04-30T12:09:09.237-05:00 level=INFO source=gpu_windows.go:183 msg="efficiency cores detected" maxEfficiencyClass=1 time=2025-04-30T12:09:09.238-05:00 level=INFO source=gpu_windows.go:214 msg="" package=0 cores=10 efficiency=8 threads=12 time=2025-04-30T12:09:09.253-05:00 level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered" time=2025-04-30T12:09:09.253-05:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="31.4 GiB" available="13.7 GiB" ``` ### OS Windows ### GPU Intel ### CPU Intel ### Ollama version 0.6.6
GiteaMirror added the bug label 2026-04-12 18:48:12 -05:00
GiteaMirror commented 2026-04-12 18:48:13 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 30, 2025):

The runner architecture has changed, there are no longer runner variants for different CPU architectures. Instead the runner dynamically loads a library to implement architecture-dependent processing. Load a model and look for a line in the logs with compiler= in it:

time=2025-04-30T16:45:32.662Z level=INFO source=ggml.go:109 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)
<!-- gh-comment-id:2842767440 --> @rick-github commented on GitHub (Apr 30, 2025): The runner architecture has changed, there are no longer runner variants for different CPU architectures. Instead the runner dynamically loads a library to implement architecture-dependent processing. Load a model and look for a line in the logs with `compiler=` in it: ``` time=2025-04-30T16:45:32.662Z level=INFO source=ggml.go:109 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc) ```
GiteaMirror commented 2026-04-12 18:48:14 -05:00
Author
Owner
Copy Link

@aaronoaks commented on GitHub (Apr 30, 2025):

Thanks for the fast response! I checked the logs and found the corresponding line:

time=2025-04-30T17:39:10.236Z level=INFO source=ggml.go:109 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc)

I'm assuming the =1 means it will be able to use the associated instruction sets?

<!-- gh-comment-id:2842856410 --> @aaronoaks commented on GitHub (Apr 30, 2025): Thanks for the fast response! I checked the logs and found the corresponding line: ``` time=2025-04-30T17:39:10.236Z level=INFO source=ggml.go:109 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc) ``` I'm assuming the `=1` means it will be able to use the associated instruction sets?
GiteaMirror commented 2026-04-12 18:48:14 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 30, 2025):

Yep.

<!-- gh-comment-id:2842922449 --> @rick-github commented on GitHub (Apr 30, 2025): Yep.
GiteaMirror commented 2026-04-12 18:48:15 -05:00
Author
Owner
Copy Link

@aaronoaks commented on GitHub (Apr 30, 2025):

Ok, thanks for the info, good to know where to check now.

<!-- gh-comment-id:2842983474 --> @aaronoaks commented on GitHub (Apr 30, 2025): Ok, thanks for the info, good to know where to check now.
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#6913
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?