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 #8741] Longer context length leads to half power usage? #67727

New Issue
Closed
opened 2026-05-04 11:29:09 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-05-04 11:29:09 -05:00
Owner
Copy Link

Originally created by @kungfu-eric on GitHub (Jan 31, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8741

What is the issue?

It seems cranking up the context length beyond the default 4096 (24000) on deepseek-r1 32b (qwen 2.5) and 70b (llama3.3) causes power usage to drop by half with corresponding throughput drop.

There's no OOM on VRAM. Others suggested this is due to paging to RAM or CPU activity. Glances says CPU is operating at ~22% and there's no disk IO wait reported.

Is there some inefficiency in Ollama?

OS

Linux

GPU

Nvidia

CPU

AMD

Ollama version

0.3.12

Originally created by @kungfu-eric on GitHub (Jan 31, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8741 ### What is the issue? It seems cranking up the context length beyond the default 4096 (24000) on deepseek-r1 32b (qwen 2.5) and 70b (llama3.3) causes power usage to drop by half with corresponding throughput drop. There's no OOM on VRAM. Others suggested this is due to paging to RAM or CPU activity. Glances says CPU is operating at ~22% and there's no disk IO wait reported. Is there some inefficiency in Ollama? ### OS Linux ### GPU Nvidia ### CPU AMD ### Ollama version 0.3.12
GiteaMirror added the bug label 2026-05-04 11:29:09 -05:00
GiteaMirror commented 2026-05-04 11:29:14 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 31, 2025):

Server logs would give some insight.

<!-- gh-comment-id:2628578818 --> @rick-github commented on GitHub (Jan 31, 2025): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) would give some insight.
GiteaMirror commented 2026-05-04 11:29:16 -05:00
Author
Owner
Copy Link

@kungfu-eric commented on GitHub (Feb 1, 2025):

Doing some reasoning, pretty sure it's memory bandwidth bound unfortunately. Closing

<!-- gh-comment-id:2628595684 --> @kungfu-eric commented on GitHub (Feb 1, 2025): Doing some reasoning, pretty sure it's memory bandwidth bound unfortunately. Closing
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#67727
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?