github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 08:59:13 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #14342] re-add OLLAMA_MAX_VRAM #45878

New Issue
Open
opened 2026-04-25 01:29:31 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-25 01:29:31 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/14342
Author: @machineonamission
Created: 2/21/2026
Status: 🔄 Open

Base: main ← Head: main

📝 Commits (5)

  • 6fd5f3e re-add OLLAMA_MAX_VRAM
  • d2abd1c Merge branch 'ollama:main' into main
  • c8edf04 Merge branch 'ollama:main' into main
  • 15f8d50 Merge branch 'umain'
  • 89ba810 Merge branch 'ollama:main' into main

📊 Changes

5 files changed (+11 additions, -4 deletions)

View changed files

📝 cmd/cmd.go (+1 -0)
📝 envconfig/config.go (+2 -0)
📝 llm/server.go (+2 -0)
📝 server/routes.go (+1 -1)
📝 server/sched.go (+5 -3)

📄 Description

OLLAMA_MAX_VRAM is a parameter that once existed to limit the maximum amount of vram it consumes. it was deprecated because "it was a hack" and "GPU_OVERHEAD" should be used instead, but I personally find utility in this (reserving a section of a server VRAM for ollama). Ollama can now, with this param set, do its "auto calculate how much to put into vram" thing, but for any vram size, instead of manually fiddling with num_gpu for every single model. It is per-gpu now, which is documented.

Tested and working on Windows 11

🔄 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/14342 **Author:** [@machineonamission](https://github.com/machineonamission) **Created:** 2/21/2026 **Status:** 🔄 Open **Base:** `main` ← **Head:** `main` --- ### 📝 Commits (5) - [`6fd5f3e`](https://github.com/ollama/ollama/commit/6fd5f3e4456d30303e417e8d5b1de15b2366a930) re-add OLLAMA_MAX_VRAM - [`d2abd1c`](https://github.com/ollama/ollama/commit/d2abd1c9f4e1d380e3beb6b773ffe5c1f84ccae2) Merge branch 'ollama:main' into main - [`c8edf04`](https://github.com/ollama/ollama/commit/c8edf04a485ae1f640f406dcd8d51870788ad5c3) Merge branch 'ollama:main' into main - [`15f8d50`](https://github.com/ollama/ollama/commit/15f8d5010ee15b87f67235f4a31bff7eb37d2d87) Merge branch 'umain' - [`89ba810`](https://github.com/ollama/ollama/commit/89ba810624eb6daa8d2caaee2436d74885512638) Merge branch 'ollama:main' into main ### 📊 Changes **5 files changed** (+11 additions, -4 deletions) <details> <summary>View changed files</summary> 📝 `cmd/cmd.go` (+1 -0) 📝 `envconfig/config.go` (+2 -0) 📝 `llm/server.go` (+2 -0) 📝 `server/routes.go` (+1 -1) 📝 `server/sched.go` (+5 -3) </details> ### 📄 Description `OLLAMA_MAX_VRAM` is a parameter that once existed to limit the maximum amount of vram it consumes. it was deprecated because "it was a hack" and "GPU_OVERHEAD" should be used instead, but I personally find utility in this (reserving a section of a server VRAM for ollama). Ollama can now, with this param set, do its "auto calculate how much to put into vram" thing, but for any vram size, instead of manually fiddling with num_gpu for every single model. It is per-gpu now, which is documented. Tested and working on Windows 11 --- <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 2026-04-25 01:29:31 -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#45878
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?