github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-05 23:53:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #14642] [MERGED] MLX: add header vendoring and remove go build tag #40633

New Issue
Closed
opened 2026-04-23 01:29:33 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-23 01:29:33 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/14642
Author: @dhiltgen
Created: 3/5/2026
Status: ✅ Merged
Merged: 3/10/2026
Merged by: @dhiltgen

Base: main ← Head: mlx_vendor

📝 Commits (4)

  • 04f6fd6 prefer rocm v6 on windows
  • 4403584 MLX: add header vendoring and remove go build tag
  • 6246157 ci: harden for flaky choco repo servers
  • f12ed61 review comments

📊 Changes

142 files changed (+5350 additions, -1064 deletions)

View changed files

📝 .github/workflows/release.yaml (+46 -6)
📝 .github/workflows/test.yaml (+55 -9)
📝 CMakeLists.txt (+94 -12)
📝 CMakePresets.json (+1 -0)
📝 Dockerfile (+17 -6)
➕ MLX_CORE_VERSION (+1 -0)
📝 docs/development.md (+68 -0)
📝 parser/parser.go (+3 -0)
📝 scripts/build_darwin.sh (+3 -3)
📝 scripts/build_windows.ps1 (+146 -59)
📝 scripts/env.sh (+8 -0)
📝 x/create/client/quantize.go (+3 -4)
➖ x/create/client/quantize_stub.go (+0 -25)
📝 x/imagegen/cache/cache.go (+0 -2)
📝 x/imagegen/cache/step.go (+0 -2)
📝 x/imagegen/cache/teacache.go (+0 -2)
📝 x/imagegen/cmd/engine/README.md (+1 -1)
📝 x/imagegen/cmd/engine/generate.go (+0 -2)
📝 x/imagegen/cmd/engine/image.go (+0 -2)
📝 x/imagegen/cmd/engine/main.go (+0 -2)

...and 80 more files

📄 Description

This switches to using a vendoring approach for the mlx-c headers so that Go can build without requiring a cmake first. This enables building the new MLX based code by default. Every time cmake runs, the headers are refreshed, so we
can easily keep them in sync when we bump mlx versions. Basic Windows and Linux support are verified.

🔄 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/14642 **Author:** [@dhiltgen](https://github.com/dhiltgen) **Created:** 3/5/2026 **Status:** ✅ Merged **Merged:** 3/10/2026 **Merged by:** [@dhiltgen](https://github.com/dhiltgen) **Base:** `main` ← **Head:** `mlx_vendor` --- ### 📝 Commits (4) - [`04f6fd6`](https://github.com/ollama/ollama/commit/04f6fd61e27c917950485015ac061bc87c082c1f) prefer rocm v6 on windows - [`4403584`](https://github.com/ollama/ollama/commit/440358459ac637d698fa4dd167d0b512d760fb7e) MLX: add header vendoring and remove go build tag - [`6246157`](https://github.com/ollama/ollama/commit/6246157fbc486051250c8148305083e78e640069) ci: harden for flaky choco repo servers - [`f12ed61`](https://github.com/ollama/ollama/commit/f12ed618111734e3037265574bbf9afdf91cb317) review comments ### 📊 Changes **142 files changed** (+5350 additions, -1064 deletions) <details> <summary>View changed files</summary> 📝 `.github/workflows/release.yaml` (+46 -6) 📝 `.github/workflows/test.yaml` (+55 -9) 📝 `CMakeLists.txt` (+94 -12) 📝 `CMakePresets.json` (+1 -0) 📝 `Dockerfile` (+17 -6) ➕ `MLX_CORE_VERSION` (+1 -0) 📝 `docs/development.md` (+68 -0) 📝 `parser/parser.go` (+3 -0) 📝 `scripts/build_darwin.sh` (+3 -3) 📝 `scripts/build_windows.ps1` (+146 -59) 📝 `scripts/env.sh` (+8 -0) 📝 `x/create/client/quantize.go` (+3 -4) ➖ `x/create/client/quantize_stub.go` (+0 -25) 📝 `x/imagegen/cache/cache.go` (+0 -2) 📝 `x/imagegen/cache/step.go` (+0 -2) 📝 `x/imagegen/cache/teacache.go` (+0 -2) 📝 `x/imagegen/cmd/engine/README.md` (+1 -1) 📝 `x/imagegen/cmd/engine/generate.go` (+0 -2) 📝 `x/imagegen/cmd/engine/image.go` (+0 -2) 📝 `x/imagegen/cmd/engine/main.go` (+0 -2) _...and 80 more files_ </details> ### 📄 Description This switches to using a vendoring approach for the mlx-c headers so that Go can build without requiring a cmake first. This enables building the new MLX based code by default. Every time cmake runs, the headers are refreshed, so we can easily keep them in sync when we bump mlx versions. Basic Windows and Linux support are verified. --- <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 2026-04-23 01:29:33 -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#40633
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?