Skip to content

All on-device models

Image to image models for iPhone and Mac (Core ML & MLX)

48 models for this task are tagged Core ML or MLX on Hugging Face, or come from Apple: 21 Core ML, 27 MLX and 0 other Apple models. The table lists the most downloaded in the last 30 days.

Last updated 2026-11-02 (UTC).

Showing 48 of 48.

ModelSizeDownloads (30 days)
mflux-community/flux2-klein-4b-mflux-q4MLXfrom black-forest-labs/FLUX.2-klein-4B4.6 GB42.2K
mflux-community/qwen-image-edit-2511-mflux-q4MLXfrom Qwen/Qwen-Image-Edit-251129 GB34.4K
mflux-community/qwen-image-edit-2511-mflux-q8MLXfrom Qwen/Qwen-Image-Edit-251137.5 GB24.9K
mflux-community/flux-1-dev-kontext-mflux-q4MLXfrom black-forest-labs/FLUX.1-Kontext-dev9.6 GB8,709
mflux-community/flux2-klein-4b-mflux-q8MLXfrom black-forest-labs/FLUX.2-klein-4B—5,430
mflux-community/flux2-klein-9b-mflux-q4MLXfrom black-forest-labs/FLUX.2-klein-9B—5,307
mflux-community/qwen-image-edit-2511-mflux-q6MLXfrom Qwen/Qwen-Image-Edit-2511—5,036
mflux-community/qwen-image-edit-2509-mflux-q8MLXfrom Qwen/Qwen-Image-Edit-2509—4,787
mflux-community/flux2-klein-9b-mflux-q8MLXfrom black-forest-labs/FLUX.2-klein-9B—3,537
mflux-community/flux-2-klein-base-4b-mflux-q4MLXfrom black-forest-labs/FLUX.2-klein-base-4B—3,382
mflux-community/flux-1-dev-kontext-mflux-q8MLXfrom black-forest-labs/FLUX.1-Kontext-dev—3,270
mflux-community/flux2-klein-9b-kv-mflux-q4MLXfrom black-forest-labs/FLUX.2-klein-9b-kv—2,428
mflux-community/flux2-klein-4b-mflux-q6MLXfrom black-forest-labs/FLUX.2-klein-4B—2,203
mflux-community/qwen-image-edit-2509-mflux-q4MLXfrom Qwen/Qwen-Image-Edit-2509—1,326
mflux-community/fibo-edit-1-5-turbo-mflux-q4MLXfrom briaai/Fibo-Edit-1.5-turbo—1,286
mlx-community/MuseTalk-1.5-fp16MLXfrom TMElyralab/MuseTalk—1,140
mflux-community/qwen-image-edit-2511-mflux-q3MLXfrom Qwen/Qwen-Image-Edit-2511—1,130
mflux-community/qwen-image-edit-2511-mflux-bf16MLXfrom Qwen/Qwen-Image-Edit-2511—1,055
mlboydaisuke/LaMa-CoreMLCore ML—1,012
mlx-community/MuseTalk-1.5-q4MLXfrom TMElyralab/MuseTalk—955
mflux-community/qwen-image-edit-2509-mflux-q3MLXfrom Qwen/Qwen-Image-Edit-2509—946
mflux-community/flux-2-klein-base-4b-mflux-q8MLXfrom black-forest-labs/FLUX.2-klein-base-4B—934
mflux-community/flux2-klein-9b-mflux-bf16MLXfrom black-forest-labs/FLUX.2-klein-9B—806
mflux-community/flux-2-klein-base-9b-mflux-q8MLXfrom black-forest-labs/FLUX.2-klein-base-9B—735
mflux-community/flux2-klein-9b-kv-mflux-bf16MLXfrom black-forest-labs/FLUX.2-klein-9b-kv—668
mflux-community/qwen-image-edit-mflux-bf16MLXfrom Qwen/Qwen-Image-Edit—609
mflux-community/qwen-image-edit-2509-mflux-bf16MLXfrom Qwen/Qwen-Image-Edit-2509—593
mflux-community/flux2-klein-9b-kv-mflux-q6MLXfrom black-forest-labs/FLUX.2-klein-9b-kv—575
mlboydaisuke/GFPGAN-CoreMLCore ML—501
LucasTakanori/orosense_latCore ML—326
mlboydaisuke/Real-ESRGAN-x4-CoreMLCore ML—188
LucasTakanori/orosense_sideCore ML—178
ModelPiper/PiperSR-2xCore ML—124
arraypress/upscale-realesrganCore ML—102
languel/genereticore-upscalers-coremlCore ML—76
ModernoTech/FaceSwap-ModelsCore ML—70
StereoMaker/stereospace-v1-0-coremlCore MLfrom prs-eth/stereospace-v1-0—52
VioletXF/manganinja-coremlCore ML—48
neonforestmist/Clover-Image-Tiny-InpaintCore MLfrom neonforestmist/Clover-Image-Tiny—40
arraypress/upscale-ncCore ML—34
heypatkay/solume-sdxl-inpainting-coremlCore MLfrom diffusers/stable-diffusion-xl-1.0-inpainting-0.1—29
kakegu/stable-diffusion-xl-1.0-inpainting-0.1-coreml-fp16Core MLfrom diffusers/stable-diffusion-xl-1.0-inpainting-0.1—28
gumka/realplksr-coremlCore ML—26
heypatkay/solume-big-lama-coremlCore MLfrom smartywu/big-lama—22
dinnerrush/AdcSR-x4-CoreMLCore MLfrom Guaishou74851/AdcSR—19
mlboydaisuke/DDColor-Tiny-CoreMLCore ML—18
arraypress/erase-lamaCore MLfrom smartywu/big-lama—17
gumka/lama-coremlCore ML—17

Reading the table

  • Downloads are Hugging Face's count for the last 30 days, not all time. Hugging Face counts requests for a model's main file, so one install can count more than once.
  • Size is the total of the repo's current files on Hugging Face, in the sizes it shows (1 GB = 1,000 MB). Some repos hold several variants of a model, so it can be larger than the one download you need; a dash means it has not been measured yet.
  • Trending is Hugging Face's own score for recent attention. "from" under a name is the model it was made from.
  • Parameters is the parameter count in the model files. A count marked * comes from the model's name (such as 27B) instead: the files gave none, or a much smaller one, as packed low-bit files can.
  • Quantization is read from the model's tags and name, not checked against the files.

Before you use one

The list comes from Hugging Face's public API and is updated daily. We don't test each model. Check the license and read the model card before you ship one in an app: some licenses limit commercial use, and "other" means the terms are on the model card. Models whose names or tags point at adult or safety-removed content are left out, and so are gated models that need a Hugging Face sign-in to download.