github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #13648] [MERGED] Add experimental MLX backend and engine with imagegen support #76607

New Issue
Closed
opened 2026-05-05 09:14:40 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-05-05 09:14:40 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/13648
Author: @dhiltgen
Created: 1/8/2026
Status: ✅ Merged
Merged: 1/9/2026
Merged by: @dhiltgen

Base: main ← Head: mlx-engine

📝 Commits (10+)

  • 3d90a54 WIP - MLX backend with gemma3
  • e16578d MLX: add cmake and go tag build toggles
  • d6e3048 add experimental image generation runtime
  • 231fe1b add experimental image generation runtime
  • 92f3354 MLX: wire up cuda build for linux
  • 65ca1f3 MLX: get dependencies correct and dedup
  • c84992d fix relative link bug in dedup
  • aa11cf3 Add darwin build and readme
  • 27b719e add go build tag for mlx dependent code and wire up build_darwin.sh
  • 0c193b5 lint cleanup

📊 Changes

123 files changed (+27490 additions, -88 deletions)

View changed files

📝 CMakeLists.txt (+59 -9)
📝 CMakePresets.json (+38 -1)
📝 Dockerfile (+35 -1)
📝 convert/convert.go (+132 -24)
📝 convert/convert_bert.go (+1 -1)
📝 convert/convert_commandr.go (+1 -1)
📝 convert/convert_deepseek2.go (+1 -1)
📝 convert/convert_deepseekocr.go (+1 -1)
📝 convert/convert_gemma.go (+1 -1)
📝 convert/convert_gemma2.go (+1 -3)
📝 convert/convert_gemma2_adapter.go (+2 -1)
📝 convert/convert_gemma3.go (+1 -3)
📝 convert/convert_gemma3n.go (+1 -1)
📝 convert/convert_gptoss.go (+1 -1)
📝 convert/convert_llama.go (+1 -1)
📝 convert/convert_llama4.go (+1 -1)
📝 convert/convert_llama_adapter.go (+5 -4)
📝 convert/convert_mistral.go (+1 -1)
📝 convert/convert_mistral_causal.go (+1 -1)
📝 convert/convert_mixtral.go (+1 -1)

...and 80 more files

📄 Description

Add experimental support for a new MLX based backend and image generation model support.

🔄 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/13648 **Author:** [@dhiltgen](https://github.com/dhiltgen) **Created:** 1/8/2026 **Status:** ✅ Merged **Merged:** 1/9/2026 **Merged by:** [@dhiltgen](https://github.com/dhiltgen) **Base:** `main` ← **Head:** `mlx-engine` --- ### 📝 Commits (10+) - [`3d90a54`](https://github.com/ollama/ollama/commit/3d90a542a31adab0209d8f53437eb3caaa06c582) WIP - MLX backend with gemma3 - [`e16578d`](https://github.com/ollama/ollama/commit/e16578d6bb76816e10db05e887207cd6ad22d090) MLX: add cmake and go tag build toggles - [`d6e3048`](https://github.com/ollama/ollama/commit/d6e30481af216cfb7c5d3849a0df953066200fe9) add experimental image generation runtime - [`231fe1b`](https://github.com/ollama/ollama/commit/231fe1b39733f557b025ddcd9faaaf2b52e165e3) add experimental image generation runtime - [`92f3354`](https://github.com/ollama/ollama/commit/92f33547e2df486d920ac4fa9e9a5a2f04859de6) MLX: wire up cuda build for linux - [`65ca1f3`](https://github.com/ollama/ollama/commit/65ca1f3954a1f92dbea5b7505ccb0cc3ba7b9099) MLX: get dependencies correct and dedup - [`c84992d`](https://github.com/ollama/ollama/commit/c84992dbdf6240b19a80c7b2fb817f14a9a8ff68) fix relative link bug in dedup - [`aa11cf3`](https://github.com/ollama/ollama/commit/aa11cf3dc15c1e4d5e882a8b402ea2c1990cd2c7) Add darwin build and readme - [`27b719e`](https://github.com/ollama/ollama/commit/27b719e4a2b6dbbec5017fa1692edfd850a6042e) add go build tag for mlx dependent code and wire up build_darwin.sh - [`0c193b5`](https://github.com/ollama/ollama/commit/0c193b51f58f9c75def7da9162bd629aa71e7d19) lint cleanup ### 📊 Changes **123 files changed** (+27490 additions, -88 deletions) <details> <summary>View changed files</summary> 📝 `CMakeLists.txt` (+59 -9) 📝 `CMakePresets.json` (+38 -1) 📝 `Dockerfile` (+35 -1) 📝 `convert/convert.go` (+132 -24) 📝 `convert/convert_bert.go` (+1 -1) 📝 `convert/convert_commandr.go` (+1 -1) 📝 `convert/convert_deepseek2.go` (+1 -1) 📝 `convert/convert_deepseekocr.go` (+1 -1) 📝 `convert/convert_gemma.go` (+1 -1) 📝 `convert/convert_gemma2.go` (+1 -3) 📝 `convert/convert_gemma2_adapter.go` (+2 -1) 📝 `convert/convert_gemma3.go` (+1 -3) 📝 `convert/convert_gemma3n.go` (+1 -1) 📝 `convert/convert_gptoss.go` (+1 -1) 📝 `convert/convert_llama.go` (+1 -1) 📝 `convert/convert_llama4.go` (+1 -1) 📝 `convert/convert_llama_adapter.go` (+5 -4) 📝 `convert/convert_mistral.go` (+1 -1) 📝 `convert/convert_mistral_causal.go` (+1 -1) 📝 `convert/convert_mixtral.go` (+1 -1) _...and 80 more files_ </details> ### 📄 Description Add experimental support for a new MLX based backend and image generation model support. --- <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 2026-05-05 09:14:40 -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#76607
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?