github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #6983] [CLOSED] Add support for CC v5 and v6+ multi-gpu cuda setups #12280

New Issue
Closed
opened 2026-04-12 23:54:08 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 23:54:08 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/6983
Author: @dhiltgen
Created: 9/26/2024
Status: ❌ Closed

Base: main ← Head: multi_variant

📝 Commits (1)

  • c7acd8e Add support for v5 and v6+ multi-gpu cuda setups

📊 Changes

3 files changed (+87 additions, -10 deletions)

View changed files

📝 discover/gpu_test.go (+56 -1)
📝 discover/types.go (+30 -8)
📝 llm/server.go (+1 -1)

📄 Description

With the new addition of cuda v12 and selecting the newer version for GPUs that support it, users with a mix of v11 only and v12 supporting GPUs weren't able to load models spanning the GPU. This refines the algorithm to use the oldest variant (v11 in this case) so that all the GPUs will be supported. This will break down if a user has a CC 5.x card and a CC 9.x card, as we don't currently have plumbing in place to understand the maximum version support.

Fixes #6930
Fixes #8066

🔄 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/6983 **Author:** [@dhiltgen](https://github.com/dhiltgen) **Created:** 9/26/2024 **Status:** ❌ Closed **Base:** `main` ← **Head:** `multi_variant` --- ### 📝 Commits (1) - [`c7acd8e`](https://github.com/ollama/ollama/commit/c7acd8e24815732c01bb704ccaf0fca7be0531a3) Add support for v5 and v6+ multi-gpu cuda setups ### 📊 Changes **3 files changed** (+87 additions, -10 deletions) <details> <summary>View changed files</summary> 📝 `discover/gpu_test.go` (+56 -1) 📝 `discover/types.go` (+30 -8) 📝 `llm/server.go` (+1 -1) </details> ### 📄 Description With the new addition of cuda v12 and selecting the newer version for GPUs that support it, users with a mix of v11 only and v12 supporting GPUs weren't able to load models spanning the GPU. This refines the algorithm to use the oldest variant (v11 in this case) so that all the GPUs will be supported. This will break down if a user has a CC 5.x card and a CC 9.x card, as we don't currently have plumbing in place to understand the maximum version support. Fixes #6930 Fixes #8066 --- <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 2026-04-12 23:54:08 -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#12280
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?