github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7557] 我的文件明明只有12GB，为什么加载的时候，却显示需要22GB显存？ #4811

New Issue
Closed
opened 2026-04-12 15:47:20 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-12 15:47:20 -05:00
Owner
Copy Link

Originally created by @fg2501 on GitHub (Nov 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7557

What is the issue?

3f5ba904-d558-4770-b3e3-ee7783694b2f
1111
我用的qwen2.5:14b模型，为什么我加载的时候，显示却是需要22GB显存呢？

OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

0.3.13

Originally created by @fg2501 on GitHub (Nov 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7557 ### What is the issue? ![3f5ba904-d558-4770-b3e3-ee7783694b2f](https://github.com/user-attachments/assets/1a2ef30a-766c-4f00-873c-0b3468238754) ![1111](https://github.com/user-attachments/assets/03dba386-b3cb-4267-9d02-07b048c3a503) 我用的qwen2.5:14b模型，为什么我加载的时候，显示却是需要22GB显存呢？ ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version 0.3.13
GiteaMirror added the question label 2026-04-12 15:47:20 -05:00
GiteaMirror commented 2026-04-12 15:47:21 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 7, 2024):

Context window.

<!-- gh-comment-id:2462258823 --> @rick-github commented on GitHub (Nov 7, 2024): [Context window](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-can-i-specify-the-context-window-size).
GiteaMirror commented 2026-04-12 15:47:22 -05:00
Author
Owner
Copy Link

@dhiltgen commented on GitHub (Nov 7, 2024):

ollama list shows the size of the model on disk. Once loaded into the GPU, more VRAM is required, which varies from model to model based on their architecture.

<!-- gh-comment-id:2463297227 --> @dhiltgen commented on GitHub (Nov 7, 2024): `ollama list` shows the size of the model on disk. Once loaded into the GPU, more VRAM is required, which varies from model to model based on their architecture.
GiteaMirror commented 2026-04-12 15:47:23 -05:00
Author
Owner
Copy Link

@fg2501 commented on GitHub (Nov 8, 2024):

ollama list 显示模型在磁盘上的大小。一旦加载到 GPU 中，就需要更多的 VRAM，这取决于它们的架构，各不相同。

好的非常感谢！

<!-- gh-comment-id:2463565450 --> @fg2501 commented on GitHub (Nov 8, 2024): > `ollama list` 显示模型在磁盘上的大小。一旦加载到 GPU 中，就需要更多的 VRAM，这取决于它们的架构，各不相同。 好的非常感谢！
GiteaMirror commented 2026-04-12 15:47:23 -05:00
Author
Owner
Copy Link

@fg2501 commented on GitHub (Nov 8, 2024):

问题已经解决，并不单单是上下文长度的问题，而是环境变量的问题，OLLAMA_NUM_PARALLEL=16，改成1以后，就好了，同时处理的信息太大所导致的显存量增加。

<!-- gh-comment-id:2463680035 --> @fg2501 commented on GitHub (Nov 8, 2024): 问题已经解决，并不单单是上下文长度的问题，而是环境变量的问题，OLLAMA_NUM_PARALLEL=16，改成1以后，就好了，同时处理的信息太大所导致的显存量增加。
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 question
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#4811
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?