[PR #363] [CLOSED] linux cuda build tag #10119

Closed
opened 2026-04-12 22:52:07 -05:00 by GiteaMirror · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/363
Author: @BruceMacD
Created: 8/16/2023
Status: Closed

Base: mainHead: brucemacd/linux-nvidia-build


📝 Commits (5)

📊 Changes

4 files changed (+21 additions, -0 deletions)

View changed files

📝 docs/development.md (+16 -0)
llm/generate_cuda.go (+3 -0)
llm/ggml-cuda.syso (+0 -0)
📝 llm/llama.go (+2 -0)

📄 Description

This change allows building a linux go binary using a cuda tag. The goal of this tag behaviour is that running go build . will still build a Go binary that works on CPU, while adding the cuda tag allows building a go binary that uses nvidia graphics cards (via cuda). This allows everyone that has been using ollama on linux to continue to build from source without any breaking changes.

To build with cuda:
go build -tags cuda .


🔄 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/363 **Author:** [@BruceMacD](https://github.com/BruceMacD) **Created:** 8/16/2023 **Status:** ❌ Closed **Base:** `main` ← **Head:** `brucemacd/linux-nvidia-build` --- ### 📝 Commits (5) - [`11bb296`](https://github.com/ollama/ollama/commit/11bb2960a7b3c0c96a6af9aa11e50899d0370d77) linux cuda build tag - [`e217127`](https://github.com/ollama/ollama/commit/e21712756b55ffbdb06222bf2884584420262810) pr feedback - [`6680a6e`](https://github.com/ollama/ollama/commit/6680a6edf3908a7b4dbdc2e1a34e1348f0366815) CPPFLAGS - [`6d70524`](https://github.com/ollama/ollama/commit/6d70524d7f62220599522a5bea972823a7bc8a93) add ggml-cuda - [`0c924dc`](https://github.com/ollama/ollama/commit/0c924dc8fbb1566e764e9f24dc7b800ea65e62f8) Update llama.go ### 📊 Changes **4 files changed** (+21 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `docs/development.md` (+16 -0) ➕ `llm/generate_cuda.go` (+3 -0) ➕ `llm/ggml-cuda.syso` (+0 -0) 📝 `llm/llama.go` (+2 -0) </details> ### 📄 Description This change allows building a linux go binary using a cuda tag. The goal of this tag behaviour is that running `go build .` will still build a Go binary that works on CPU, while adding the cuda tag allows building a go binary that uses nvidia graphics cards (via cuda). This allows everyone that has been using ollama on linux to continue to build from source without any breaking changes. To build with cuda: `go build -tags cuda .` --- <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 22:52:07 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#10119