github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #10773] [MERGED] llm: Use first layer as memory buffer in estimation #39231

New Issue
Closed
opened 2026-04-22 23:53:42 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-22 23:53:42 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/10773
Author: @jessegross
Created: 5/19/2025
Status: ✅ Merged
Merged: 5/19/2025
Merged by: @jessegross

Base: main ← Head: jessegross/layers

📝 Commits (1)

  • 1fcfeab llm: Use first layer as memory buffer in estimation

📊 Changes

1 file changed (+6 additions, -7 deletions)

View changed files

📝 llm/memory.go (+6 -7)

📄 Description

This is a partial revert of 0478d44 "Fixed over vram allcation dure to small initial layer sizes."

Previously we used the size of the first layer as an extra reserved amount of space to buffer our memory estimates. The above commit changed this to use the largest layer. However, this had performance impacts on more models than the original commit was trying to fix.

There is just a heuristic without an ideal solution so this goes back to the historic behavior.

Fixes: #10765, #10756, #10752, #10726

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/ollama/ollama/pull/10773 **Author:** [@jessegross](https://github.com/jessegross) **Created:** 5/19/2025 **Status:** ✅ Merged **Merged:** 5/19/2025 **Merged by:** [@jessegross](https://github.com/jessegross) **Base:** `main` ← **Head:** `jessegross/layers` --- ### 📝 Commits (1) - [`1fcfeab`](https://github.com/ollama/ollama/commit/1fcfeabc0cb46e0274d5d5b196b6d5c0c16a7d84) llm: Use first layer as memory buffer in estimation ### 📊 Changes **1 file changed** (+6 additions, -7 deletions) <details> <summary>View changed files</summary> 📝 `llm/memory.go` (+6 -7) </details> ### 📄 Description This is a partial revert of 0478d44 "Fixed over vram allcation dure to small initial layer sizes." Previously we used the size of the first layer as an extra reserved amount of space to buffer our memory estimates. The above commit changed this to use the largest layer. However, this had performance impacts on more models than the original commit was trying to fix. There is just a heuristic without an ideal solution so this goes back to the historic behavior. Fixes: #10765, #10756, #10752, #10726 --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-04-22 23:53:42 -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 pull-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#39231
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?