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 #5458] Enable Row Split Support #3414

New Issue
Open
opened 2026-04-12 14:03:22 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 14:03:22 -05:00
Owner
Copy Link

Originally created by @datacrystals on GitHub (Jul 3, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5458

For multi-gpu setups, enabling row-split can have very significant performance improvements. On my machine with 3xP40, I was able to get a speedup from ~3t/s to ~10t/s.

Unfortunately, it doesn't look like there's any way to get this flag (-sm row) down to llama.cpp, which should otherwise support this.

Wondering if it's possible to either add a way to pass along flags to llama.cpp, or to have an option which enables this flag. Given the massive performance improvement that it brings though, it seems like this should be possible, and not very much work to implement.

I saw that passing along flags was discussed here, but was unfortunately not implemented.

Originally created by @datacrystals on GitHub (Jul 3, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5458 For multi-gpu setups, enabling row-split can have very significant performance improvements. On my machine with 3xP40, I was able to get a speedup from ~3t/s to ~10t/s. Unfortunately, it doesn't look like there's any way to get this flag (`-sm row`) down to llama.cpp, which should otherwise support this. Wondering if it's possible to either add a way to pass along flags to llama.cpp, or to have an option which enables this flag. Given the massive performance improvement that it brings though, it seems like this should be possible, and not very much work to implement. I saw that passing along flags was discussed [here](https://github.com/ollama/ollama/pull/4120#issuecomment-2094747527), but was unfortunately not implemented.
GiteaMirror added the feature request label 2026-04-12 14:03:22 -05:00
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#3414
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?