github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #1730] MLX backend #47497

New Issue
Open
opened 2026-04-28 03:56:24 -05:00 by GiteaMirror · 95 comments
GiteaMirror commented 2026-04-28 03:56:24 -05:00
Owner
Copy Link

Originally created by @ageorgios on GitHub (Dec 27, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1730

Can ollama be converted to use MLX from Apple as backend for the models ?

Originally created by @ageorgios on GitHub (Dec 27, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1730 Can ollama be converted to use MLX from Apple as backend for the models ?
GiteaMirror added the feature request label 2026-04-28 03:56:25 -05:00
GiteaMirror commented 2026-04-28 03:56:26 -05:00
Author
Owner
Copy Link

@Josecodesalot commented on GitHub (Dec 31, 2023):

This Please!

<!-- gh-comment-id:1872909090 --> @Josecodesalot commented on GitHub (Dec 31, 2023): This Please!
GiteaMirror commented 2026-04-28 03:56:26 -05:00
Author
Owner
Copy Link

@easp commented on GitHub (Jan 2, 2024):

What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet.

<!-- gh-comment-id:1874541941 --> @easp commented on GitHub (Jan 2, 2024): What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet.
GiteaMirror commented 2026-04-28 03:56:27 -05:00
Author
Owner
Copy Link

@KernelBypass commented on GitHub (Jan 10, 2024):

Found these benchmarks:
https://medium.com/@andreask_75652/benchmarking-apples-mlx-vs-llama-cpp-bbbebdc18416

Seems like MLX is indeed slower than the llama.cpp masterpiece, at least for now. I did not verify though.

<!-- gh-comment-id:1884193340 --> @KernelBypass commented on GitHub (Jan 10, 2024): Found these benchmarks: https://medium.com/@andreask_75652/benchmarking-apples-mlx-vs-llama-cpp-bbbebdc18416 Seems like MLX is indeed slower than the llama.cpp masterpiece, at least for now. I did not verify though.
GiteaMirror commented 2026-04-28 03:56:27 -05:00
Author
Owner
Copy Link

@Edu126 commented on GitHub (Jan 23, 2024):

This would be very nice!
and not only for text generation, Image/Multimodal would be boosted too.

<!-- gh-comment-id:1905227427 --> @Edu126 commented on GitHub (Jan 23, 2024): This would be very nice! and not only for text generation, Image/Multimodal would be boosted too.
GiteaMirror commented 2026-04-28 03:56:27 -05:00
Author
Owner
Copy Link

@JimmyLv commented on GitHub (Apr 20, 2024):

someone made this https://github.com/kspviswa/PyOMlx

<!-- gh-comment-id:2067654729 --> @JimmyLv commented on GitHub (Apr 20, 2024): someone made this https://github.com/kspviswa/PyOMlx
GiteaMirror commented 2026-04-28 03:56:28 -05:00
Author
Owner
Copy Link

@magnusviri commented on GitHub (May 4, 2024):

Ollama is awesome and does so many things and some of us want to play with mlx models.

<!-- gh-comment-id:2093998058 --> @magnusviri commented on GitHub (May 4, 2024): Ollama is awesome and does so many things and some of us want to play with mlx models.
GiteaMirror commented 2026-04-28 03:56:28 -05:00
Author
Owner
Copy Link

@angelo-cortez commented on GitHub (May 30, 2024):

bump

<!-- gh-comment-id:2138596675 --> @angelo-cortez commented on GitHub (May 30, 2024): bump
GiteaMirror commented 2026-04-28 03:56:28 -05:00
Author
Owner
Copy Link

@mxyng commented on GitHub (May 30, 2024):

Commenting here to say we're aware of MLX. I've been working on a prototype but I can't give an ETA at for MLX support at this time

<!-- gh-comment-id:2138701258 --> @mxyng commented on GitHub (May 30, 2024): Commenting here to say we're aware of MLX. I've been working on a prototype but I can't give an ETA at for MLX support at this time
GiteaMirror commented 2026-04-28 03:56:29 -05:00
Author
Owner
Copy Link

@qdrddr commented on GitHub (Jun 28, 2024):

Related to this Apple CoreML support to utilize Apple Neural Engine (ANE) alongside GPU & CPU:
https://github.com/ollama/ollama/issues/3898

<!-- gh-comment-id:2197640182 --> @qdrddr commented on GitHub (Jun 28, 2024): Related to this Apple CoreML support to utilize Apple Neural Engine (ANE) alongside GPU & CPU: https://github.com/ollama/ollama/issues/3898
GiteaMirror commented 2026-04-28 03:56:29 -05:00
Author
Owner
Copy Link

@ibehnam commented on GitHub (Aug 20, 2024):

Any updates on this? MLX is now faster than Llama.cpp on Mac.

<!-- gh-comment-id:2299217776 --> @ibehnam commented on GitHub (Aug 20, 2024): Any updates on this? MLX is now faster than Llama.cpp on Mac.
GiteaMirror commented 2026-04-28 03:56:29 -05:00
Author
Owner
Copy Link

@garhbod commented on GitHub (Aug 21, 2024):

Any progress @mxyng? Is this a seperate project that other could contribute to?

<!-- gh-comment-id:2300018379 --> @garhbod commented on GitHub (Aug 21, 2024): Any progress @mxyng? Is this a seperate project that other could contribute to?
GiteaMirror commented 2026-04-28 03:56:29 -05:00
Author
Owner
Copy Link

@nicarq commented on GitHub (Aug 29, 2024):

it would be awesome. MLX is moving really fast and it would make sense that it would be the best tool long-term to run models on Apple's hardware.

<!-- gh-comment-id:2319102840 --> @nicarq commented on GitHub (Aug 29, 2024): it would be awesome. MLX is moving really fast and it would make sense that it would be the best tool long-term to run models on Apple's hardware.
GiteaMirror commented 2026-04-28 03:56:30 -05:00
Author
Owner
Copy Link

@parthpat12 commented on GitHub (Sep 9, 2024):

Please add support for MLX! Any update @mxyng?

<!-- gh-comment-id:2338711590 --> @parthpat12 commented on GitHub (Sep 9, 2024): Please add support for MLX! Any update @mxyng?
GiteaMirror commented 2026-04-28 03:56:30 -05:00
Author
Owner
Copy Link

@ivanfioravanti commented on GitHub (Sep 15, 2024):

MLX support would be awesome!!!

<!-- gh-comment-id:2351620475 --> @ivanfioravanti commented on GitHub (Sep 15, 2024): MLX support would be awesome!!!
GiteaMirror commented 2026-04-28 03:56:30 -05:00
Author
Owner
Copy Link

@czzarr commented on GitHub (Sep 15, 2024):

indeed it would be!

<!-- gh-comment-id:2351739080 --> @czzarr commented on GitHub (Sep 15, 2024): indeed it would be!
GiteaMirror commented 2026-04-28 03:56:31 -05:00
Author
Owner
Copy Link

@vietvudanh commented on GitHub (Oct 2, 2024):

MLX would support vision models too.

<!-- gh-comment-id:2387842211 --> @vietvudanh commented on GitHub (Oct 2, 2024): MLX would support vision models too.
GiteaMirror commented 2026-04-28 03:56:31 -05:00
Author
Owner
Copy Link

@Bigsy commented on GitHub (Oct 9, 2024):

Seems the new MLX backend in LMStudio is providing some real benefits especially in regards to memory consumption. Would be great to get support in Ollama.

<!-- gh-comment-id:2401916894 --> @Bigsy commented on GitHub (Oct 9, 2024): Seems the new MLX backend in LMStudio is providing some real benefits especially in regards to memory consumption. Would be great to get support in Ollama.
GiteaMirror commented 2026-04-28 03:56:32 -05:00
Author
Owner
Copy Link

@hg0428 commented on GitHub (Oct 10, 2024):

I have been testing the MLX backend in LM Studio, and I have found it to be on average 40% faster for inference than Ollama using the same exact settings with the same model at the same precision.
I am using M3 Max 36GB memory.

<!-- gh-comment-id:2405262489 --> @hg0428 commented on GitHub (Oct 10, 2024): I have been testing the MLX backend in LM Studio, and I have found it to be on average 40% faster for inference than Ollama using the same exact settings with the same model at the same precision. I am using M3 Max 36GB memory.
GiteaMirror commented 2026-04-28 03:56:33 -05:00
Author
Owner
Copy Link

@robbiemu commented on GitHub (Oct 18, 2024):

I have been testing the MLX backend in LM Studio, and I have found it to be on average 40% faster for inference than Ollama using the same exact settings with the same model at the same precision. I am using M3 Max 36GB memory.

I've seen numbers, admittedly a couple months ago, around 20% faster. Can you share a bit more - what models/context/settings? what iogpu.wired_limit_mb? etc

20% is still 20% more than Im doing currently :D

I'm not sure how good of an idea it is to have Ollama add a lot of features only available to some people .. but it does have some NVIDIA exclusive (or nvidia vs cpu only) stuff at least.

<!-- gh-comment-id:2423291403 --> @robbiemu commented on GitHub (Oct 18, 2024): > I have been testing the MLX backend in LM Studio, and I have found it to be on average 40% faster for inference than Ollama using the same exact settings with the same model at the same precision. I am using M3 Max 36GB memory. I've seen numbers, admittedly a couple months ago, around 20% faster. Can you share a bit more - what models/context/settings? what iogpu.wired_limit_mb? etc 20% is still 20% more than Im doing currently :D I'm not sure how good of an idea it is to have Ollama add a lot of features only available to some people .. but it does have some NVIDIA exclusive (or nvidia vs cpu only) stuff at least.
GiteaMirror commented 2026-04-28 03:56:34 -05:00
Author
Owner
Copy Link

@CharafChnioune commented on GitHub (Oct 20, 2024):

So any plans to add mlx like lmstudio? Mlx supports multi model and is faster now. Llama.cpp is sort of death since the stoped vision

<!-- gh-comment-id:2425207478 --> @CharafChnioune commented on GitHub (Oct 20, 2024): So any plans to add mlx like lmstudio? Mlx supports multi model and is faster now. Llama.cpp is sort of death since the stoped vision
GiteaMirror commented 2026-04-28 03:56:34 -05:00
Author
Owner
Copy Link

@twalderman commented on GitHub (Oct 23, 2024):

Given that there are less options with MLX for models but the ones that are available are good workhorses, it would be a huge benefit to have Ollama support mlx. As others have stated, LM studio supports MLX and the performance is great however Ollama still supports a wider range of templates and potentially upcoming support for more sampler options. Having one solution is ideal for Apple Silicon.

<!-- gh-comment-id:2432489080 --> @twalderman commented on GitHub (Oct 23, 2024): Given that there are less options with MLX for models but the ones that are available are good workhorses, it would be a huge benefit to have Ollama support mlx. As others have stated, LM studio supports MLX and the performance is great however Ollama still supports a wider range of templates and potentially upcoming support for more sampler options. Having one solution is ideal for Apple Silicon.
GiteaMirror commented 2026-04-28 03:56:35 -05:00
Author
Owner
Copy Link

@ice6 commented on GitHub (Oct 30, 2024):

it is good to support this :) most important keep ollama steady and fast!

<!-- gh-comment-id:2445683055 --> @ice6 commented on GitHub (Oct 30, 2024): it is good to support this :) most important keep `ollama` steady and fast!
GiteaMirror commented 2026-04-28 03:56:35 -05:00
Author
Owner
Copy Link

@nercone-dev commented on GitHub (Oct 31, 2024):

On the MacBook Air (Apple M3 Normal), it is now faster.
For example, it was faster even when running 13B Codellama.
This is probably a technology that can be optimized for Apple silicon (especially M3 and later).
So it would be better to implement it.

https://github.com/user-attachments/assets/6aadc898-1c62-43c8-91d6-8c2308db603c

https://github.com/user-attachments/assets/e65e373f-a2bd-4089-82d0-66c3cbab4db5

Model

MacBook Air 13-inch (2024, M3)

CPU

Apple M3

Memory (RAM)

16GB

<!-- gh-comment-id:2449908124 --> @nercone-dev commented on GitHub (Oct 31, 2024): On the MacBook Air (Apple M3 Normal), it is now **faster**. For example, it was **faster even when running 13B Codellama**. This is probably a technology that can be optimized for Apple silicon (especially M3 and later). So it would be better to implement it. https://github.com/user-attachments/assets/6aadc898-1c62-43c8-91d6-8c2308db603c https://github.com/user-attachments/assets/e65e373f-a2bd-4089-82d0-66c3cbab4db5 ## Model MacBook Air 13-inch (2024, M3) ## CPU Apple M3 ## Memory (RAM) 16GB
GiteaMirror commented 2026-04-28 03:56:35 -05:00
Author
Owner
Copy Link

@hg0428 commented on GitHub (Oct 31, 2024):

On my hardware, MLX runs an average of 40% faster than Llama.cpp (actual percentage varies from 38%-42%. 40% is the average over many tests).

<!-- gh-comment-id:2449987822 --> @hg0428 commented on GitHub (Oct 31, 2024): On my hardware, MLX runs an average of 40% faster than Llama.cpp (actual percentage varies from 38%-42%. 40% is the average over many tests).
GiteaMirror commented 2026-04-28 03:56:36 -05:00
Author
Owner
Copy Link

@twalderman commented on GitHub (Oct 31, 2024):

What are people using for mlx local serving? What implementation is best and worth implementing in ollama?

<!-- gh-comment-id:2450041164 --> @twalderman commented on GitHub (Oct 31, 2024): What are people using for mlx local serving? What implementation is best and worth implementing in ollama?
GiteaMirror commented 2026-04-28 03:56:36 -05:00
Author
Owner
Copy Link

@hg0428 commented on GitHub (Oct 31, 2024):

What are people using for mlx local serving? What implementation is best and worth implementing in ollama?

The developers of LM Studio have created a wrapper around MLX that makes it super simple. They used it to transition LM Studio from supporting only Llama.cpp as a backend to being able to support both Llama.cpp and MLX.

https://github.com/lmstudio-ai/mlx-engine

<!-- gh-comment-id:2450112199 --> @hg0428 commented on GitHub (Oct 31, 2024): > What are people using for mlx local serving? What implementation is best and worth implementing in ollama? The developers of LM Studio have created a wrapper around MLX that makes it super simple. They used it to transition LM Studio from supporting only Llama.cpp as a backend to being able to support both Llama.cpp and MLX. https://github.com/lmstudio-ai/mlx-engine
GiteaMirror commented 2026-04-28 03:56:37 -05:00
Author
Owner
Copy Link

@ahmetkca commented on GitHub (Nov 9, 2024):

What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet.

I hav just tried LM Studio's new MLX backend and you can see 11+ tokens per second improvement for same model. The model in question was qwen2.5:7b-instruct-q8_0 from 70~ tokens per second to 81~ tokens per second

<!-- gh-comment-id:2466166189 --> @ahmetkca commented on GitHub (Nov 9, 2024): > What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet. I hav just tried LM Studio's new MLX backend and you can see 11+ tokens per second improvement for same model. The model in question was qwen2.5:7b-instruct-q8_0 from 70~ tokens per second to 81~ tokens per second
GiteaMirror commented 2026-04-28 03:56:37 -05:00
Author
Owner
Copy Link

@ahmetkca commented on GitHub (Nov 9, 2024):

What are people using for mlx local serving? What implementation is best and worth implementing in ollama?

The developers of LM Studio have created a wrapper around MLX that makes it super simple. They used it to transition LM Studio from supporting only Llama.cpp as a backend to being able to support both Llama.cpp and MLX.

https://github.com/lmstudio-ai/mlx-engine

https://github.com/ollama/ollama/issues/1730#issuecomment-2466166189

<!-- gh-comment-id:2466166358 --> @ahmetkca commented on GitHub (Nov 9, 2024): > > What are people using for mlx local serving? What implementation is best and worth implementing in ollama? > > The developers of LM Studio have created a wrapper around MLX that makes it super simple. They used it to transition LM Studio from supporting only Llama.cpp as a backend to being able to support both Llama.cpp and MLX. > > https://github.com/lmstudio-ai/mlx-engine https://github.com/ollama/ollama/issues/1730#issuecomment-2466166189
GiteaMirror commented 2026-04-28 03:56:37 -05:00
Author
Owner
Copy Link

@ahmetkca commented on GitHub (Nov 9, 2024):

What are people using for mlx local serving? What implementation is best and worth implementing in ollama?

The developers of LM Studio have created a wrapper around MLX that makes it super simple. They used it to transition LM Studio from supporting only Llama.cpp as a backend to being able to support both Llama.cpp and MLX.

https://github.com/lmstudio-ai/mlx-engine

This is actually wrapper around MLX's mlx-lm python package. So perhaps, Ollama team can do better by completely bypassing python?

<!-- gh-comment-id:2466167374 --> @ahmetkca commented on GitHub (Nov 9, 2024): > > What are people using for mlx local serving? What implementation is best and worth implementing in ollama? > > The developers of LM Studio have created a wrapper around MLX that makes it super simple. They used it to transition LM Studio from supporting only Llama.cpp as a backend to being able to support both Llama.cpp and MLX. > > https://github.com/lmstudio-ai/mlx-engine This is actually wrapper around MLX's mlx-lm python package. So perhaps, Ollama team can do better by completely bypassing python?
GiteaMirror commented 2026-04-28 03:56:37 -05:00
Author
Owner
Copy Link

@logiota commented on GitHub (Nov 10, 2024):

What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet.

llama3.2 is at least 64% faster with MLX on M4! Just got mine :)

<!-- gh-comment-id:2466964333 --> @logiota commented on GitHub (Nov 10, 2024): > What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet. llama3.2 is at least 64% faster with MLX on M4! Just got mine :)
GiteaMirror commented 2026-04-28 03:56:38 -05:00
Author
Owner
Copy Link

@Idmon commented on GitHub (Nov 18, 2024):

Are we gonna see MLX support?

I think now with the new M4 Max chips it's a great time to support it.

<!-- gh-comment-id:2482833792 --> @Idmon commented on GitHub (Nov 18, 2024): Are we gonna see MLX support? I think now with the new M4 Max chips it's a great time to support it.
GiteaMirror commented 2026-04-28 03:56:38 -05:00
Author
Owner
Copy Link

@ljgeneral commented on GitHub (Nov 19, 2024):

Followed, looking forward to ollama support, I will try LM Studio first

<!-- gh-comment-id:2484585563 --> @ljgeneral commented on GitHub (Nov 19, 2024): Followed, looking forward to ollama support, I will try LM Studio first
GiteaMirror commented 2026-04-28 03:56:38 -05:00
Author
Owner
Copy Link

@bhupesh-sf commented on GitHub (Nov 22, 2024):

Want to follow to see its support natively

<!-- gh-comment-id:2494723422 --> @bhupesh-sf commented on GitHub (Nov 22, 2024): Want to follow to see its support natively
GiteaMirror commented 2026-04-28 03:56:38 -05:00
Author
Owner
Copy Link

@NashvilleBrandon commented on GitHub (Nov 27, 2024):

The people need this.

<!-- gh-comment-id:2502327666 --> @NashvilleBrandon commented on GitHub (Nov 27, 2024): The people need this.
GiteaMirror commented 2026-04-28 03:56:39 -05:00
Author
Owner
Copy Link

@bakaburg1 commented on GitHub (Dec 4, 2024):

Up!

<!-- gh-comment-id:2516666080 --> @bakaburg1 commented on GitHub (Dec 4, 2024): Up!
GiteaMirror commented 2026-04-28 03:56:39 -05:00
Author
Owner
Copy Link

@baoduy commented on GitHub (Dec 5, 2024):

1 more vote for this month be 👍👍👍

<!-- gh-comment-id:2519630493 --> @baoduy commented on GitHub (Dec 5, 2024): 1 more vote for this month be 👍👍👍
GiteaMirror commented 2026-04-28 03:56:40 -05:00
Author
Owner
Copy Link

@elfriscia commented on GitHub (Dec 5, 2024):

1 more vote for this month be 👍👍👍

Wish granted. One more 👍
This feature is native on Lm Studio and there’s a big difference.

<!-- gh-comment-id:2519981038 --> @elfriscia commented on GitHub (Dec 5, 2024): > 1 more vote for this month be 👍👍👍 Wish granted. One more 👍 This feature is native on Lm Studio and there’s a big difference.
GiteaMirror commented 2026-04-28 03:56:41 -05:00
Author
Owner
Copy Link

@vietvudanh commented on GitHub (Dec 6, 2024):

Well, I ended up using MLX directly. Just wish they supports json output mode.

<!-- gh-comment-id:2521933938 --> @vietvudanh commented on GitHub (Dec 6, 2024): Well, I ended up using MLX directly. Just wish they supports json output mode.
GiteaMirror commented 2026-04-28 03:56:42 -05:00
Author
Owner
Copy Link

@smsmatt commented on GitHub (Dec 7, 2024):

Would appear to be a big win for Ollama to put this feature in.

<!-- gh-comment-id:2525062383 --> @smsmatt commented on GitHub (Dec 7, 2024): Would appear to be a big win for Ollama to put this feature in.
GiteaMirror commented 2026-04-28 03:56:43 -05:00
Author
Owner
Copy Link

@cryptedx commented on GitHub (Dec 8, 2024):

Well, this would be a big win for Ollama!

<!-- gh-comment-id:2526319015 --> @cryptedx commented on GitHub (Dec 8, 2024): Well, this would be a big win for Ollama!
GiteaMirror commented 2026-04-28 03:56:43 -05:00
Author
Owner
Copy Link

@hg0428 commented on GitHub (Dec 8, 2024):

Llama.cpp is exploring Apple Silicon ANE support, which not even MLX has. If implemented properly, that could make Llama.cpp significantly faster than MLX.

<!-- gh-comment-id:2526319832 --> @hg0428 commented on GitHub (Dec 8, 2024): Llama.cpp is exploring Apple Silicon ANE support, which not even MLX has. If implemented properly, that could make Llama.cpp significantly faster than MLX.
GiteaMirror commented 2026-04-28 03:56:43 -05:00
Author
Owner
Copy Link

@ivanfioravanti commented on GitHub (Dec 8, 2024):

All models must be created adhoc for ANE. Moreover ANE is faster if you have few GPU cores otherwise noSent from the road.Il giorno 8 dic 2024, alle ore 19:44, Hudson Gouge @.***> ha scritto:
Llama.cpp is exploring Apple Silicon ANE support, which not even MLX has. If implemented properly, that could make Llama.cpp significantly faster than MLX.

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: @.***>

<!-- gh-comment-id:2526325495 --> @ivanfioravanti commented on GitHub (Dec 8, 2024): All models must be created adhoc for ANE. Moreover ANE is faster if you have few GPU cores otherwise noSent from the road.Il giorno 8 dic 2024, alle ore 19:44, Hudson Gouge ***@***.***> ha scritto: Llama.cpp is exploring Apple Silicon ANE support, which not even MLX has. If implemented properly, that could make Llama.cpp significantly faster than MLX. —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: ***@***.***>
GiteaMirror commented 2026-04-28 03:56:44 -05:00
Author
Owner
Copy Link

@hg0428 commented on GitHub (Dec 8, 2024):

All models must be created adhoc for ANE. Moreover ANE is faster if you have few GPU cores otherwise no.

The ANE is like the GPU, but specialized for AI. The model files themselves need not be changed. You can run anything on it just like the GPU, thanks the new API that Apple released. Previously, you had to use CoreML; now, you can access it directly. The ANE alone can get performance roughly equivalent to the M4 Max GPU, which is quite good. Combined with GPU, it can result in a significant performance boost.

<!-- gh-comment-id:2526328586 --> @hg0428 commented on GitHub (Dec 8, 2024): > All models must be created adhoc for ANE. Moreover ANE is faster if you have few GPU cores otherwise no. The ANE is like the GPU, but specialized for AI. The model files themselves need not be changed. You can run anything on it just like the GPU, thanks the new API that Apple released. Previously, you had to use CoreML; now, you can access it directly. The ANE alone can get performance roughly equivalent to the M4 Max GPU, which is quite good. Combined with GPU, it can result in a significant performance boost.
GiteaMirror commented 2026-04-28 03:56:45 -05:00
Author
Owner
Copy Link

@elfriscia commented on GitHub (Dec 8, 2024):

The troll downvoting is frustrated for not being able to have a MacBook and will reply anything to contradict this statement.

<!-- gh-comment-id:2526371663 --> @elfriscia commented on GitHub (Dec 8, 2024): The troll downvoting is frustrated for not being able to have a MacBook and will reply anything to contradict this statement.
GiteaMirror commented 2026-04-28 03:56:45 -05:00
Author
Owner
Copy Link

@vietvudanh commented on GitHub (Dec 9, 2024):

Well, soon I guess: Ollama's post

<!-- gh-comment-id:2526781705 --> @vietvudanh commented on GitHub (Dec 9, 2024): Well, soon I guess: [Ollama's post](https://x.com/ollama/status/1865238754052485293)
GiteaMirror commented 2026-04-28 03:56:45 -05:00
Author
Owner
Copy Link

@iamhenry commented on GitHub (Dec 9, 2024):

image

<!-- gh-comment-id:2528935945 --> @iamhenry commented on GitHub (Dec 9, 2024): ![image](https://github.com/user-attachments/assets/2d0b000d-a651-4ca3-8653-33546c2f66d3)
GiteaMirror commented 2026-04-28 03:56:45 -05:00
Author
Owner
Copy Link

@integrate-your-mind commented on GitHub (Dec 11, 2024):

If this happens by Monday I will be so happy.

<!-- gh-comment-id:2533769453 --> @integrate-your-mind commented on GitHub (Dec 11, 2024): If this happens by Monday I will be so happy.
GiteaMirror commented 2026-04-28 03:56:46 -05:00
Author
Owner
Copy Link

@rheinardkorf commented on GitHub (Dec 17, 2024):

It did not happen Monday. 😅

<!-- gh-comment-id:2548402958 --> @rheinardkorf commented on GitHub (Dec 17, 2024): It did not happen Monday. 😅
GiteaMirror commented 2026-04-28 03:56:46 -05:00
Author
Owner
Copy Link

@loganyc1 commented on GitHub (Dec 18, 2024):

Did this ship?

<!-- gh-comment-id:2550157227 --> @loganyc1 commented on GitHub (Dec 18, 2024): Did this ship?
GiteaMirror commented 2026-04-28 03:56:48 -05:00
Author
Owner
Copy Link

@MS00-GitIt commented on GitHub (Dec 18, 2024):

Reading all this has been fun. I can't wait to see this issue a marked as "Closed".

<!-- gh-comment-id:2551746553 --> @MS00-GitIt commented on GitHub (Dec 18, 2024): Reading all this has been fun. I can't wait to see this issue a marked as "Closed".
GiteaMirror commented 2026-04-28 03:56:49 -05:00
Author
Owner
Copy Link

@zhaopengme commented on GitHub (Dec 28, 2024):

Waiting.⏳(๑˃̵ᴗ˂̵)

<!-- gh-comment-id:2564252416 --> @zhaopengme commented on GitHub (Dec 28, 2024): Waiting.⏳(๑˃̵ᴗ˂̵)
GiteaMirror commented 2026-04-28 03:56:50 -05:00
Author
Owner
Copy Link

@dalisoft commented on GitHub (Dec 30, 2024):

Reading all this has been fun. I can't wait to see this issue a marked as "Closed".

Better if Closed as completed not as Closed as not planned 😄

<!-- gh-comment-id:2565240109 --> @dalisoft commented on GitHub (Dec 30, 2024): > Reading all this has been fun. I can't wait to see this issue a marked as "Closed". Better if Closed as completed not as Closed as not planned 😄
GiteaMirror commented 2026-04-28 03:56:51 -05:00
Author
Owner
Copy Link

@Swich1987 commented on GitHub (Jan 4, 2025):

Looking forward for that support 🤞

<!-- gh-comment-id:2570394625 --> @Swich1987 commented on GitHub (Jan 4, 2025): Looking forward for that support 🤞
GiteaMirror commented 2026-04-28 03:56:51 -05:00
Author
Owner
Copy Link

@jeffhaskin commented on GitHub (Jan 16, 2025):

What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet.

Update 1 year later:

  • MLX is 25.02% faster than GGUF on my M1 Air 16G
  • MLX is 41.01% faster than GGUF on my M2 Pro 32G.

Model: Llama 3.1 8b Instruct Q4
Platform: LM Studio

On the M1 Air, even Phi4-14b is getting usable speeds (7.06 tok/sec MLX vs 5 tok/sec GGUF), which puts it just into the usability range for me.

On the M2 Pro 32G, I'm getting similar speeds with Qwen2.5 32B (mlx), which is great.

<!-- gh-comment-id:2596850995 --> @jeffhaskin commented on GitHub (Jan 16, 2025): > What do you hope to gain from this? I don't think MLX is faster for inference, at least not yet. Update 1 year later: - MLX is 25.02% faster than GGUF on my M1 Air 16G - MLX is 41.01% faster than GGUF on my M2 Pro 32G. Model: Llama 3.1 8b Instruct Q4 Platform: LM Studio On the M1 Air, even **Phi4-14b** is getting usable speeds (7.06 tok/sec MLX vs 5 tok/sec GGUF), which puts it just into the usability range for me. On the M2 Pro 32G, I'm getting similar speeds with Qwen2.5 32B (mlx), which is great.
GiteaMirror commented 2026-04-28 03:56:52 -05:00
Author
Owner
Copy Link

@iamhenry commented on GitHub (Jan 16, 2025):

monday is coming up 😅

<!-- gh-comment-id:2597063939 --> @iamhenry commented on GitHub (Jan 16, 2025): monday is coming up 😅
GiteaMirror commented 2026-04-28 03:56:52 -05:00
Author
Owner
Copy Link

@schneipk commented on GitHub (Jan 22, 2025):

I'm so stoked ! OllaMLXa <3

<!-- gh-comment-id:2606573948 --> @schneipk commented on GitHub (Jan 22, 2025): I'm so stoked ! OllaMLXa <3
GiteaMirror commented 2026-04-28 03:56:53 -05:00
Author
Owner
Copy Link

@Unicorndy commented on GitHub (Jan 24, 2025):

Is it happening soon? 🤞

<!-- gh-comment-id:2612624324 --> @Unicorndy commented on GitHub (Jan 24, 2025): Is it happening soon? 🤞
GiteaMirror commented 2026-04-28 03:56:54 -05:00
Author
Owner
Copy Link

@Byte1122 commented on GitHub (Jan 31, 2025):

I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate!

<!-- gh-comment-id:2626240230 --> @Byte1122 commented on GitHub (Jan 31, 2025): I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate!
GiteaMirror commented 2026-04-28 03:56:54 -05:00
Author
Owner
Copy Link

@bhupesh-sf commented on GitHub (Feb 1, 2025):

here we go https://github.com/ollama/ollama/pull/8490

<!-- gh-comment-id:2628824035 --> @bhupesh-sf commented on GitHub (Feb 1, 2025): here we go [https://github.com/ollama/ollama/pull/8490](https://github.com/ollama/ollama/pull/8490)
GiteaMirror commented 2026-04-28 03:56:54 -05:00
Author
Owner
Copy Link

@TheurgicDuke771 commented on GitHub (Feb 14, 2025):

I see the PR https://github.com/ollama/ollama/pull/8490 is now closed. Can we expect this in next stable release?

<!-- gh-comment-id:2658249000 --> @TheurgicDuke771 commented on GitHub (Feb 14, 2025): I see the PR https://github.com/ollama/ollama/pull/8490 is now closed. Can we expect this in next stable release?
GiteaMirror commented 2026-04-28 03:56:55 -05:00
Author
Owner
Copy Link

@kconner commented on GitHub (Feb 18, 2025):

#8490 was superseded by #9118, which seems worth watching for progress.

<!-- gh-comment-id:2664773683 --> @kconner commented on GitHub (Feb 18, 2025): #8490 was superseded by #9118, which seems worth watching for progress.
GiteaMirror commented 2026-04-28 03:56:56 -05:00
Author
Owner
Copy Link

@neolee commented on GitHub (Feb 22, 2025):

So do we have any plan and/or schedule for MLX support?

<!-- gh-comment-id:2675952617 --> @neolee commented on GitHub (Feb 22, 2025): So do we have any plan and/or schedule for MLX support?
GiteaMirror commented 2026-04-28 03:56:56 -05:00
Author
Owner
Copy Link

@HadiCherkaoui commented on GitHub (Feb 23, 2025):

Has it arrived?

<!-- gh-comment-id:2676992045 --> @HadiCherkaoui commented on GitHub (Feb 23, 2025): Has it arrived?
GiteaMirror commented 2026-04-28 03:56:56 -05:00
Author
Owner
Copy Link

@robwilkes commented on GitHub (Feb 26, 2025):

Models converted to MLX format are ~20% faster than the same models in GGUF format on my M4 MacBook Pro

<!-- gh-comment-id:2683694097 --> @robwilkes commented on GitHub (Feb 26, 2025): Models converted to MLX format are ~20% faster than the same models in GGUF format on my M4 MacBook Pro
GiteaMirror commented 2026-04-28 03:56:57 -05:00
Author
Owner
Copy Link

@Caojunisstudying commented on GitHub (Mar 14, 2025):

I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate!

However, when I use the MLX model, my Macmini takes up much more memory, so I have to abandon MLX, because if cpu performance is good enough，I would rather spend memory on building models with higher precision , I don’t know how Ollama will be running the mlx model in the future，hope better results.

<!-- gh-comment-id:2723469034 --> @Caojunisstudying commented on GitHub (Mar 14, 2025): > I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate! However, when I use the MLX model, my Macmini takes up much more memory, so I have to abandon MLX, because if cpu performance is good enough，I would rather spend memory on building models with higher precision , I don’t know how Ollama will be running the mlx model in the future，hope better results.
GiteaMirror commented 2026-04-28 03:56:57 -05:00
Author
Owner
Copy Link

@robwilkes commented on GitHub (Mar 14, 2025):

I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate!

However, when I use the MLX model, my Macmini takes up much more memory, so I have to abandon MLX, because if cpu performance is good enough，I would rather spend memory on building models with higher precision , I don’t know how Ollama will be running the mlx model in the future，hope better results.

Are you sure you're comparing the same quantization, context size, etc, like for like?

The memory usage is the same for me, maybe even a tiny bit better as Mac can manage the memory slightly better with MLX, however it's mostly negligible / identical.

The filesizes are roughly equivalent and therefore the memory utilisation is roughly equivalent.

<!-- gh-comment-id:2723745580 --> @robwilkes commented on GitHub (Mar 14, 2025): > > I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate! > > However, when I use the MLX model, my Macmini takes up much more memory, so I have to abandon MLX, because if cpu performance is good enough，I would rather spend memory on building models with higher precision , I don’t know how Ollama will be running the mlx model in the future，hope better results. Are you sure you're comparing the same quantization, context size, etc, like for like? The memory usage is the same for me, maybe even a tiny bit better as Mac can manage the memory slightly better with MLX, however it's mostly negligible / identical. The filesizes are roughly equivalent and therefore the memory utilisation is roughly equivalent.
GiteaMirror commented 2026-04-28 03:56:58 -05:00
Author
Owner
Copy Link

@Caojunisstudying commented on GitHub (Mar 14, 2025):

I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate!

However, when I use the MLX model, my Macmini takes up much more memory, so I have to abandon MLX, because if cpu performance is good enough，I would rather spend memory on building models with higher precision , I don’t know how Ollama will be running the mlx model in the future，hope better results.

Are you sure you're comparing the same quantization, context size, etc, like for like?

The memory usage is the same for me, maybe even a tiny bit better as Mac can manage the memory slightly better with MLX, however it's mostly negligible / identical.

The filesizes are roughly equivalent and therefore the memory utilisation is roughly equivalent.

like qwq-32b Q4 on ollama vs qwq-32b 4bit on LMs, my macmini has just 24G memory which of very high load in such scenario, the result is that ollama run hard but finish the answer, the LMs lead my mac to crash and have to restart. I can use ollama and docker, ragflow normally, but crash with LMs because of ram out again on 14b models.
so I think ollama in running on cpu and ram of 24G, but when mlx model use gpu to accelerate, part of the ram must be allocated to gpu, so the ram gets not enought。

<!-- gh-comment-id:2723821551 --> @Caojunisstudying commented on GitHub (Mar 14, 2025): > > > I am using LM Studio and Ollama. The MLX models on LM Studio are way faster. I hope I can switch entirely to Ollama, but with LM Studio. Anyway, love to support this. If Ollama need donations or what so ever, love to donate! > > > > > > However, when I use the MLX model, my Macmini takes up much more memory, so I have to abandon MLX, because if cpu performance is good enough，I would rather spend memory on building models with higher precision , I don’t know how Ollama will be running the mlx model in the future，hope better results. > > Are you sure you're comparing the same quantization, context size, etc, like for like? > > The memory usage is the same for me, maybe even a tiny bit better as Mac can manage the memory slightly better with MLX, however it's mostly negligible / identical. > > The filesizes are roughly equivalent and therefore the memory utilisation is roughly equivalent. like qwq-32b Q4 on ollama vs qwq-32b 4bit on LMs, my macmini has just 24G memory which of very high load in such scenario, the result is that ollama run hard but finish the answer, the LMs lead my mac to crash and have to restart. I can use ollama and docker, ragflow normally, but crash with LMs because of ram out again on 14b models. so I think ollama in running on cpu and ram of 24G, but when mlx model use gpu to accelerate, part of the ram must be allocated to gpu, so the ram gets not enought。
GiteaMirror commented 2026-04-28 03:56:58 -05:00
Author
Owner
Copy Link

@arty-hlr commented on GitHub (Mar 14, 2025):

MLX does not need to preallocate memory for the context size, so you should be able to run models with their full context size with better memory utilisation. @Caojunisstudying the whole 24G are not available to you/to the GPU, only 2/3 of those by default. Did you change the limit with sudo sysctl iogpu.wired_limit_mb=XXXXX?

<!-- gh-comment-id:2724001299 --> @arty-hlr commented on GitHub (Mar 14, 2025): MLX does not need to preallocate memory for the context size, so you should be able to run models with their full context size with better memory utilisation. @Caojunisstudying the whole 24G are not available to you/to the GPU, only 2/3 of those by default. Did you change the limit with `sudo sysctl iogpu.wired_limit_mb=XXXXX`?
GiteaMirror commented 2026-04-28 03:56:59 -05:00
Author
Owner
Copy Link

@Caojunisstudying commented on GitHub (Mar 14, 2025):

MLX does not need to preallocate memory for the context size, so you should be able to run models with their full context size with better memory utilisation. @Caojunisstudying the whole 24G are not available to you/to the GPU, only 2/3 of those by default. Did you change the limit with sudo sysctl iogpu.wired_limit_mb=XXXXX?

that may be the key, My LMs keep using the default settings, I will change it and try again, Thanks a lot for your suggestion.

<!-- gh-comment-id:2724033122 --> @Caojunisstudying commented on GitHub (Mar 14, 2025): > MLX does not need to preallocate memory for the context size, so you should be able to run models with their full context size with better memory utilisation. [@Caojunisstudying](https://github.com/Caojunisstudying) the whole 24G are not available to you/to the GPU, only 2/3 of those by default. Did you change the limit with `sudo sysctl iogpu.wired_limit_mb=XXXXX`? that may be the key, My LMs keep using the default settings, I will change it and try again, Thanks a lot for your suggestion.
GiteaMirror commented 2026-04-28 03:56:59 -05:00
Author
Owner
Copy Link

@anurmatov commented on GitHub (Mar 16, 2025):

this gpu mem limit parameter is a game changer, although it's worth keeping in mind that it's a runtime thing and resets on restarts. i've created a script that automatically sets the desired value even after restarts, might be helpful especially for headless setups

<!-- gh-comment-id:2727243272 --> @anurmatov commented on GitHub (Mar 16, 2025): this gpu mem limit parameter is a game changer, although it's worth keeping in mind that it's a runtime thing and resets on restarts. i've created a [script](https://github.com/anurmatov/mac-studio-server) that automatically sets the desired value even after restarts, might be helpful especially for headless setups
GiteaMirror commented 2026-04-28 03:57:00 -05:00
Author
Owner
Copy Link

@easp commented on GitHub (Mar 16, 2025):

man sysctl.conf

<!-- gh-comment-id:2727522114 --> @easp commented on GitHub (Mar 16, 2025): `man sysctl.conf`
GiteaMirror commented 2026-04-28 03:57:00 -05:00
Author
Owner
Copy Link

@zengqingfu1442 commented on GitHub (Mar 23, 2025):

Add mlx-vlm backend also.

<!-- gh-comment-id:2746318869 --> @zengqingfu1442 commented on GitHub (Mar 23, 2025): Add mlx-vlm backend also.
GiteaMirror commented 2026-04-28 03:57:00 -05:00
Author
Owner
Copy Link

@matthieuHenocque commented on GitHub (May 9, 2025):

Lack of MLX support is the only reason I don't use Ollama. In some cases MLX Q8 is 20% faster than GGUF, and memory usage is better handled.

So please, add MLX support to Ollama

<!-- gh-comment-id:2866224973 --> @matthieuHenocque commented on GitHub (May 9, 2025): Lack of MLX support is the only reason I don't use Ollama. In some cases MLX Q8 is 20% faster than GGUF, and memory usage is better handled. So please, add MLX support to Ollama
GiteaMirror commented 2026-04-28 03:57:01 -05:00
Author
Owner
Copy Link

@vdbwim commented on GitHub (Jun 3, 2025):

Is there a target date to get MLX supported?

<!-- gh-comment-id:2935749999 --> @vdbwim commented on GitHub (Jun 3, 2025): Is there a target date to get MLX supported?
GiteaMirror commented 2026-04-28 03:57:03 -05:00
Author
Owner
Copy Link

@sxy-trans-n commented on GitHub (Jun 4, 2025):

https://github.com/Trans-N-ai/swama

High-performance MLX-based LLM inference engine for macOS with native Swift implementation

<!-- gh-comment-id:2939422223 --> @sxy-trans-n commented on GitHub (Jun 4, 2025): https://github.com/Trans-N-ai/swama High-performance MLX-based LLM inference engine for macOS with native Swift implementation
GiteaMirror commented 2026-04-28 03:57:03 -05:00
Author
Owner
Copy Link

@FelikZ commented on GitHub (Jun 23, 2025):

https://github.com/Trans-N-ai/swama

It's a next level fast indeed. Tried yesterday with Qwen3 30B, mind blowing. Would be nice if it get support for Mistral models too.

benchmarks are crazy:

Image

<!-- gh-comment-id:2995385618 --> @FelikZ commented on GitHub (Jun 23, 2025): > https://github.com/Trans-N-ai/swama It's a next level fast indeed. Tried yesterday with Qwen3 30B, mind blowing. Would be nice if it get support for Mistral models too. [benchmarks are crazy](https://github.com/Trans-N-ai/swama/issues/15#issuecomment-2961338474): ![Image](https://github.com/user-attachments/assets/e924d4d2-e95b-4fdf-ae23-5f3e36bef748)
GiteaMirror commented 2026-04-28 03:57:04 -05:00
Author
Owner
Copy Link

@openSourcerer9000 commented on GitHub (Jul 19, 2025):

Seems ollamas mac userbase is withering away along with this PR
Draft MLX go backend for new engine by dhiltgen · Pull Request #9118 · ollama/ollama https://share.google/snsqjHmkYOqBecvLf

<!-- gh-comment-id:3092531958 --> @openSourcerer9000 commented on GitHub (Jul 19, 2025): Seems ollamas mac userbase is withering away along with this PR Draft MLX go backend for new engine by dhiltgen · Pull Request #9118 · ollama/ollama https://share.google/snsqjHmkYOqBecvLf
GiteaMirror commented 2026-04-28 03:57:04 -05:00
Author
Owner
Copy Link

@zhaopengme commented on GitHub (Aug 4, 2025):

hi.Have there been any recent developments?

<!-- gh-comment-id:3149075662 --> @zhaopengme commented on GitHub (Aug 4, 2025): hi.Have there been any recent developments?
GiteaMirror commented 2026-04-28 03:57:05 -05:00
Author
Owner
Copy Link

@HKiOnline commented on GitHub (Aug 5, 2025):

I'm using LM Studio at the moment as it has a MLX backend. The performance differences are very clear. I would love to use Ollama instead.

<!-- gh-comment-id:3156318397 --> @HKiOnline commented on GitHub (Aug 5, 2025): I'm using LM Studio at the moment as it has a MLX backend. The performance differences are very clear. I would love to use Ollama instead.
GiteaMirror commented 2026-04-28 03:57:05 -05:00
Author
Owner
Copy Link

@BradKML commented on GitHub (Aug 15, 2025):

@HKiOnline got anything similar that is FOSS AND MLX-compatible?

<!-- gh-comment-id:3191266779 --> @BradKML commented on GitHub (Aug 15, 2025): @HKiOnline got anything similar that is FOSS AND MLX-compatible?
GiteaMirror commented 2026-04-28 03:57:06 -05:00
Author
Owner
Copy Link

@CharafChnioune commented on GitHub (Aug 15, 2025):

GTA 6 or mlx support? Bets are on

Verzonden vanuit Outlook voor iOShttps://aka.ms/o0ukef

Van: Brad @.>
Verzonden: Friday, August 15, 2025 1:09:09 PM
Aan: ollama/ollama @.>
CC: Charaf @.>; Comment @.>
Onderwerp: Re: [ollama/ollama] MLX backend (Issue #1730)

[https://avatars.githubusercontent.com/u/58927531?s=20&v=4]BradKML left a comment (ollama/ollama#1730)https://github.com/ollama/ollama/issues/1730#issuecomment-3191266779

@HKiOnlinehttps://github.com/HKiOnline got anything similar that is FOSS AND MLX-compatible?

—
Reply to this email directly, view it on GitHubhttps://github.com/ollama/ollama/issues/1730#issuecomment-3191266779, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A45DQFIFM6HFXWEDL2FHSVD3NW5VLAVCNFSM6AAAAABBEW4KFSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCOJRGI3DMNZXHE.
You are receiving this because you commented.Message ID: @.***>

<!-- gh-comment-id:3191270735 --> @CharafChnioune commented on GitHub (Aug 15, 2025): GTA 6 or mlx support? Bets are on Verzonden vanuit Outlook voor iOS<https://aka.ms/o0ukef> ________________________________ Van: Brad ***@***.***> Verzonden: Friday, August 15, 2025 1:09:09 PM Aan: ollama/ollama ***@***.***> CC: Charaf ***@***.***>; Comment ***@***.***> Onderwerp: Re: [ollama/ollama] MLX backend (Issue #1730) [https://avatars.githubusercontent.com/u/58927531?s=20&v=4]BradKML left a comment (ollama/ollama#1730)<https://github.com/ollama/ollama/issues/1730#issuecomment-3191266779> @HKiOnline<https://github.com/HKiOnline> got anything similar that is FOSS AND MLX-compatible? — Reply to this email directly, view it on GitHub<https://github.com/ollama/ollama/issues/1730#issuecomment-3191266779>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A45DQFIFM6HFXWEDL2FHSVD3NW5VLAVCNFSM6AAAAABBEW4KFSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCOJRGI3DMNZXHE>. You are receiving this because you commented.Message ID: ***@***.***>
GiteaMirror commented 2026-04-28 03:57:06 -05:00
Author
Owner
Copy Link

@mkozjak commented on GitHub (Aug 15, 2025):

@HKiOnline got anything similar that is FOSS AND MLX-compatible?

mlx-omni-server, as some say, might be the best option for us.

<!-- gh-comment-id:3191280185 --> @mkozjak commented on GitHub (Aug 15, 2025): > [@HKiOnline](https://github.com/HKiOnline) got anything similar that is FOSS AND MLX-compatible? mlx-omni-server, as some say, might be the best option for us.
GiteaMirror commented 2026-04-28 03:57:07 -05:00
Author
Owner
Copy Link

@curious-boy-007 commented on GitHub (Sep 14, 2025):

@ageorgios @mkozjak @Josecodesalot
You might want to take a look at this MLX + GGUF compatible CLI tool:

<!-- gh-comment-id:3289057693 --> @curious-boy-007 commented on GitHub (Sep 14, 2025): @ageorgios @mkozjak @Josecodesalot You might want to take a look at this MLX + GGUF compatible CLI tool: - https://github.com/NexaAI/nexa-sdk
GiteaMirror commented 2026-04-28 03:57:07 -05:00
Author
Owner
Copy Link

@mkozjak commented on GitHub (Sep 14, 2025):

@ageorgios @mkozjak @Josecodesalot You might want to take a look at this MLX + GGUF compatible CLI tool:

Doesn't work at all in combination with Zed for me.

<!-- gh-comment-id:3289290697 --> @mkozjak commented on GitHub (Sep 14, 2025): > [@ageorgios](https://github.com/ageorgios) [@mkozjak](https://github.com/mkozjak) [@Josecodesalot](https://github.com/Josecodesalot) You might want to take a look at this MLX + GGUF compatible CLI tool: > > * https://github.com/NexaAI/nexa-sdk Doesn't work at all in combination with Zed for me.
GiteaMirror commented 2026-04-28 03:57:08 -05:00
Author
Owner
Copy Link

@curious-boy-007 commented on GitHub (Sep 14, 2025):

@mkozjak Please try this version for macOS installer
https://github.com/NexaAI/nexa-sdk?tab=readme-ov-file#macos
Also woud you please let me know the error log for Zed editor? In macOS build-in terminal, should work.

<!-- gh-comment-id:3289322749 --> @curious-boy-007 commented on GitHub (Sep 14, 2025): @mkozjak Please try this version for macOS installer https://github.com/NexaAI/nexa-sdk?tab=readme-ov-file#macos Also woud you please let me know the error log for Zed editor? In macOS build-in terminal, should work.
GiteaMirror commented 2026-04-28 03:57:10 -05:00
Author
Owner
Copy Link

@ericcurtin commented on GitHub (Oct 13, 2025):

In Docker Model Runner we've put effort into putting all our code in one central place to make it easier for people to contribute. Please star, fork and contribute (especially please contribute an mlx backend):

https://github.com/docker/model-runner

We have vulkan support. You can pull models from Docker Hub, Huggingface or any other OCI registry. You can also push models to Docker Hub or any other OCI registry.

<!-- gh-comment-id:3399441810 --> @ericcurtin commented on GitHub (Oct 13, 2025): In Docker Model Runner we've put effort into putting all our code in one central place to make it easier for people to contribute. Please star, fork and contribute (especially please contribute an mlx backend): https://github.com/docker/model-runner We have vulkan support. You can pull models from Docker Hub, Huggingface or any other OCI registry. You can also push models to Docker Hub or any other OCI registry.
GiteaMirror commented 2026-04-28 03:57:10 -05:00
Author
Owner
Copy Link

@Globolo001 commented on GitHub (Dec 20, 2025):

Any updates on this feature?
Very surprised it takes THIS long as LMStudio been having it for a while now (but I guess thats just the curse of OpenSource)

<!-- gh-comment-id:3677351791 --> @Globolo001 commented on GitHub (Dec 20, 2025): Any updates on this feature? Very surprised it takes THIS long as LMStudio been having it for a while now (but I guess thats just the curse of OpenSource)
GiteaMirror commented 2026-04-28 03:57:10 -05:00
Author
Owner
Copy Link

@jeffhaskin commented on GitHub (Jan 17, 2026):

Haven't touched ollama in more than a year because it doesn't support mlx. use to be the core of my system.

mlx is significantly faster on my M2 macbook than gguyf.

Our love was sweet, but it is ended. Farewell, sweet Juliette.

<!-- gh-comment-id:3764411035 --> @jeffhaskin commented on GitHub (Jan 17, 2026): Haven't touched ollama in more than a year because it doesn't support mlx. use to be the core of my system. mlx is _significantly_ faster on my M2 macbook than gguyf. Our love was sweet, but it is ended. Farewell, sweet Juliette.
GiteaMirror commented 2026-04-28 03:57:11 -05:00
Author
Owner
Copy Link

@TomLucidor commented on GitHub (Jan 19, 2026):

My current choice, they seemed to be moving very fast https://github.com/cubist38/mlx-openai-server
(as for something more conventional, Ramalama or Jan seemed good as a FOSS universal adapter compared to LMStudio?)

<!-- gh-comment-id:3766358745 --> @TomLucidor commented on GitHub (Jan 19, 2026): My current choice, they seemed to be moving very fast https://github.com/cubist38/mlx-openai-server (as for something more conventional, Ramalama or Jan seemed good as a FOSS universal adapter compared to LMStudio?)
GiteaMirror commented 2026-04-28 03:57:11 -05:00
Author
Owner
Copy Link

@Globolo001 commented on GitHub (Jan 19, 2026):

Awwwww maaaannnn.
I feel like in theory Claude Opus should just be able look at the mlx docs. Check the open source LMStudio MLX adapter
for reference and implement it in an afternoon.

Is it because just because a lack of trying, a promising work in progress, just proofing to be way harder than I naively think.
Or has an architectural roadblock been hit that effectively does not allow OLLAMA to use anything other than llama cpp

I could find two active branches. But no info what they actually try to achieve. Apart from the MLX in the branch name. (Also why don’t branches have a feature description attribute in git. So annoying just trying to figure out the intention from branch and commit names)
https://github.com/ollama/ollama/tree/mlx-gpu-cd

https://github.com/ollama/ollama/tree/mxyng/next-mlx

Also the main readme seems to allow for a custom build with mlx.
Building with MLX (experimental)

However not sure wether this is also for pulling mlx models. Or only building models using mlx.
But only did very shallow digging. If there is a better discussion I'd love to get some insights :)

<!-- gh-comment-id:3766887571 --> @Globolo001 commented on GitHub (Jan 19, 2026): Awwwww maaaannnn. I feel like in theory Claude Opus should just be able look at the mlx docs. Check the open source [LMStudio MLX adapter](https://github.com/lmstudio-ai/mlx-engine) for reference and implement it in an afternoon. Is it because just because a lack of trying, a promising work in progress, just proofing to be way harder than I naively think. Or has an architectural roadblock been hit that effectively does not allow OLLAMA to use anything other than llama cpp I could find two active branches. But no info what they actually try to achieve. Apart from the MLX in the branch name. (Also why don’t branches have a feature description attribute in git. So annoying just trying to figure out the intention from branch and commit names) https://github.com/ollama/ollama/tree/mlx-gpu-cd https://github.com/ollama/ollama/tree/mxyng/next-mlx Also the main readme seems to allow for a custom build with mlx. _Building with MLX (experimental)_ However not sure wether this is also for pulling mlx models. Or only building models using mlx. But only did very shallow digging. If there is a better discussion I'd love to get some insights :)
GiteaMirror commented 2026-04-28 03:57:11 -05:00
Author
Owner
Copy Link

@Jingyuan-Zheng commented on GitHub (Mar 4, 2026):

Is it currently supported to run MLX models in Ollama? I haven't found a configuration method or an option to enable it.

<!-- gh-comment-id:3997184473 --> @Jingyuan-Zheng commented on GitHub (Mar 4, 2026): Is it currently supported to run MLX models in Ollama? I haven't found a configuration method or an option to enable it.
GiteaMirror commented 2026-04-28 03:57:12 -05:00
Author
Owner
Copy Link

@qdrddr commented on GitHub (Mar 4, 2026):

I think it's supported now. But I can't find documentationon how to use it.
https://github.com/ollama/ollama/pull/13648

<!-- gh-comment-id:3997697443 --> @qdrddr commented on GitHub (Mar 4, 2026): I think it's supported now. But I can't find documentationon how to use it. https://github.com/ollama/ollama/pull/13648
GiteaMirror commented 2026-04-28 03:57:12 -05:00
Author
Owner
Copy Link

@jackbravo commented on GitHub (Mar 4, 2026):

And seems it is just for some usecases:

Yes, our goal is to add text and embedding models. We're starting with imagegen since that's a brand new capability for Ollama, and once we get it working well, then we'll start tackling other models. We're expecting a lot of churn in the backend code as we flesh things out and refine how we want the engine to work.

<!-- gh-comment-id:3999358674 --> @jackbravo commented on GitHub (Mar 4, 2026): And seems it is just for some usecases: > Yes, our goal is to add text and embedding models. We're starting with imagegen since that's a brand new capability for Ollama, and once we get it working well, then we'll start tackling other models. We're expecting a lot of churn in the backend code as we flesh things out and refine how we want the engine to work.
GiteaMirror commented 2026-04-28 03:57:13 -05:00
Author
Owner
Copy Link

@Byte1122 commented on GitHub (Mar 4, 2026):

Yes it is supported, guys just use ai to discover it. Not long ago this was added, also you need to enable it in the confit or modelfile

33ee7168ba

<!-- gh-comment-id:4000636808 --> @Byte1122 commented on GitHub (Mar 4, 2026): Yes it is supported, guys just use ai to discover it. Not long ago this was added, also you need to enable it in the confit or modelfile https://github.com/ollama/ollama/commit/33ee7168ba1e16c813b52dc2c9417efa1e2e9f20
GiteaMirror commented 2026-04-28 03:57:13 -05:00
Author
Owner
Copy Link

@huyz commented on GitHub (Mar 5, 2026):

Is MLX support still experimental and does it require special compilation?
I'm looking for out-of-the-box support

<!-- gh-comment-id:4001935681 --> @huyz commented on GitHub (Mar 5, 2026): Is MLX support still experimental and does it require special compilation? I'm looking for out-of-the-box support
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 feature 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#47497
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?