跳到正文

全部端侧模型

适用于 iPhone 和 Mac 的图像生成模型（Core ML 与 MLX）

该任务共有 102 个模型在 Hugging Face 上标为 Core ML 或 MLX，或由 Apple 发布：Core ML 15 个、MLX 86 个、其他 Apple 模型 1 个。下表列出最近 30 天下载量最高的模型。

最后更新：2026-11-02（UTC）。

显示 100 / 100 个。

模型大小下载量（30 天）
Runpod/FLUX.2-klein-4B-mflux-4bitMLX基于 black-forest-labs/FLUX.2-klein-4B4.6 GB6.9万
mflux-community/z-image-turbo-mflux-q4MLX基于 Tongyi-MAI/Z-Image-Turbo5.9 GB3万
mflux-community/z-image-turbo-mflux-q8MLX基于 Tongyi-MAI/Z-Image-Turbo11 GB2.5万
mlx-community/Qwen-Image-2.1-mflux-q4MLX基于 Qwen/Qwen-Image-2.19.6 GB2.2万
mflux-community/qwen-image-2-1-turbo-mflux-q4MLX基于 Qwen/Qwen-Image-2.1-Turbo10 GB1.6万
mlx-community/Qwen-Image-2512-4bitMLX基于 Qwen/Qwen-Image-251225.9 GB1.6万
mflux-community/qwen-image-2-1-turbo-mflux-q8MLX基于 Qwen/Qwen-Image-2.1-Turbo17.8 GB1.5万
mflux-community/flux-1-schnell-mflux-q4MLX基于 black-forest-labs/FLUX.1-schnell9.6 GB1.4万
abenzerps/Qwen-Image-2.1-Turbo-QuantizedMLX基于 Qwen/Qwen-Image-2.1-Turbo90 GB1.2万
mflux-community/qwen-image-2-1-mflux-q4MLX基于 Qwen/Qwen-Image-2.110.6 GB1万
mflux-community/krea-2-turbo-mflux-q8MLX基于 krea/Krea-2-Turbo22.2 GB9,697
mflux-community/qwen-image-2512-mflux-q8MLX基于 Qwen/Qwen-Image-251236.1 GB9,370
mflux-community/krea-2-turbo-mflux-q4MLX基于 krea/Krea-2-Turbo15.8 GB7,514
JoyFusionAI/Qwen-Image-2.1-MLX-4bitMLX基于 Qwen/Qwen-Image-2.120.5 GB7,467
mflux-community/flux-3-image-klein-mflux-q8MLX基于 black-forest-labs/flux-3-image-klein9.9 GB7,242
mflux-community/qwen-image-2512-mflux-q4MLX基于 Qwen/Qwen-Image-2512—5,604
mlx-community/Qwen-Image-2512-8bitMLX基于 Qwen/Qwen-Image—5,202
mflux-community/qwen-image-2-1-controlnet-union-mflux-q4MLX基于 Qwen/Qwen-Image-2.1—4,969
mflux-community/z-image-turbo-mflux-bf16MLX基于 Tongyi-MAI/Z-Image-Turbo—4,251
mflux-community/flux-3-image-klein-mflux-bf16MLX基于 black-forest-labs/flux-3-image-klein—3,780
mflux-community/flux-1-dev-fill-mflux-q4MLX基于 black-forest-labs/FLUX.1-Fill-dev—3,495
argmaxinc/mlx-FLUX.1-schnell-4bit-quantizedMLX—3,441
JoyFusionAI/Qwen-Image-2.1-MLX-8bitMLX基于 Qwen/Qwen-Image-2.1—3,303
mflux-community/qwen-image-2-1-mflux-bf16MLX基于 Qwen/Qwen-Image-2.1—3,193
mflux-community/flux-3-image-klein-mflux-q4MLX基于 black-forest-labs/flux-3-image-klein—3,064
mflux-community/qwen-image-2-1-mflux-q8MLX基于 Qwen/Qwen-Image-2.1—3,060
mflux-community/flux-1-schnell-mflux-q8MLX基于 black-forest-labs/FLUX.1-schnell—2,458
joeynyc/Studio-Image-Turbo-MLX-6bit-from-Qwen-2.1MLX基于 Qwen/Qwen-Image-2.1-Turbo—2,271
ddalcu/Qwen-Image-2.1-MLX-Serve-8bitMLX基于 Qwen/Qwen-Image-2.1—1,994
MLXBits/krea-2-mlx-q8MLX基于 krea/Krea-2-Turbo—1,937
mlx-community/Qwen-Image-2512-6bitMLX基于 Qwen/Qwen-Image-2512—1,909
mflux-community/qwen-image-2-1-turbo-mflux-q6MLX基于 Qwen/Qwen-Image-2.1-Turbo—1,821
mflux-community/moss-image-sft-mflux-q4MLX基于 OpenMOSS-Team/MOSS-Image-SFT—1,796
joeynyc/Studio-Image-Turbo-MLX-4bit-from-Qwen-2.1MLX基于 Qwen/Qwen-Image-2.1-Turbo—1,728
mflux-community/moss-image-sft-mflux-q8MLX基于 OpenMOSS-Team/MOSS-Image-SFT—1,718
mflux-community/qwen-image-mflux-bf16MLX基于 Qwen/Qwen-Image—1,703
mh73772/krea-2-turbo-mflux-q8-4stepMLX基于 krea/Krea-2-Turbo—1,680
mflux-community/z-image-turbo-mflux-q6MLX基于 Tongyi-MAI/Z-Image-Turbo—1,666
vanch007/FLUX.3-Image-Klein-mlx-4bitMLX—1,433
mflux-community/flux-1-dev-fill-mflux-q8MLX基于 black-forest-labs/FLUX.1-Fill-dev—1,418
mflux-community/ming-image-0-1-mflux-q3MLX基于 inclusionAI/Ming-Image-0.1-Design—1,359
joeynyc/Ming-Image-0.1-Design-mflux-q8-te5MLX基于 inclusionAI/Ming-Image-0.1-Design—1,314
mflux-community/qwen-image-2512-mflux-q6MLX基于 Qwen/Qwen-Image-2512—1,241
NovaeonStudio/qwen-image-2.1-mflux-q8MLX基于 Qwen/Qwen-Image-2.1—1,199
vanch007/FLUX.3-Image-Klein-mlx-bf16MLX—1,161
ddalcu/Qwen-Image-2.1-MLX-Serve-4bitMLX基于 Qwen/Qwen-Image-2.1—1,159
SirSahOl/Qwen-Image-mlx-16bitMLX基于 Qwen/Qwen-Image—1,156
mflux-community/fibo-2-0-edit-turbo-mflux-q4MLX基于 briaai/Fibo-2.0-edit-turbo—1,156
mflux-community/krea-2-raw-mflux-q4MLX基于 krea/Krea-2-Raw—1,123
mflux-community/flux-3-image-klein-mflux-q6MLX基于 black-forest-labs/flux-3-image-klein—1,121
mflux-community/qwen-image-mflux-q4MLX基于 Qwen/Qwen-Image—1,100
deepsweet/Z-Image-Turbo-6B-MLX-Q8MLX基于 Tongyi-MAI/Z-Image-Turbo—1,099
mflux-community/qwen-image-mflux-q6MLX基于 Qwen/Qwen-Image—1,054
mflux-community/fibo-lite-mflux-q4MLX基于 briaai/Fibo-lite—1,025
mflux-community/ernie-image-base-mflux-q4MLX基于 baidu/ERNIE-Image—1,002
MLXBits/krea-2-mlx-q4MLX基于 krea/Krea-2-Turbo—991
ddalcu/Qwen-Image-2.1-Turbo-MLX-Serve-8bitMLX基于 Qwen/Qwen-Image-2.1-Turbo—945
argmaxinc/mlx-FLUX.1-schnellMLX—924
mflux-community/qwen-image-2512-mflux-bf16MLX基于 Qwen/Qwen-Image-2512—920
mflux-community/z-image-turbo-mflux-q5MLX基于 Tongyi-MAI/Z-Image-Turbo—906
prism-ml/bonsai-image-ternary-4B-mlx-2bitMLX基于 prism-ml/bonsai-image-ternary-4B-unpacked—893
mflux-community/flux-1-dev-mflux-q8MLX基于 black-forest-labs/FLUX.1-dev—889
mflux-community/ernie-image-turbo-mflux-q8MLX基于 baidu/ERNIE-Image-Turbo—887
argmaxinc/mlx-stable-diffusion-3.5-largeMLX基于 stabilityai/stable-diffusion-3.5-large—884
mflux-community/qwen-image-mflux-q8MLX基于 Qwen/Qwen-Image—870
mflux-community/ideogram-4-mflux-q3MLX基于 ideogram-ai/ideogram-4-fp8—863
mflux-community/z-image-base-mflux-q8MLX基于 Tongyi-MAI/Z-Image—853
vanch007/FLUX.3-Image-Klein-mlx-8bitMLX—816
joeynyc/Ming-Image-0.1-Design-mflux-q8-te8MLX基于 inclusionAI/Ming-Image-0.1-Design—805
mflux-community/fibo-2-0-dev-mflux-q8MLX基于 briaai/Fibo-2.0-dev—798
mflux-community/qwen-image-2-1-turbo-mflux-bf16MLX基于 Qwen/Qwen-Image-2.1-Turbo—795
mflux-community/flux-1-dev-controlnet-canny-mflux-q8MLX基于 black-forest-labs/FLUX.1-dev—778
mflux-community/qwen-image-2-1-mflux-q3MLX基于 Qwen/Qwen-Image-2.1—767
SirSahOl/Qwen-Image-mlx-4bitMLX基于 Qwen/Qwen-Image—752
mflux-community/flux-1-dev-mflux-q4MLX基于 black-forest-labs/FLUX.1-dev—751
deepsweet/Z-Image-Turbo-6B-MLX-Q4MLX基于 Tongyi-MAI/Z-Image-Turbo—693
mflux-community/iris-3b-turbo-mflux-q4MLX基于 speridlabs/iris-3b—662
mlx-community/Qwen-Image-2512-3bitMLX基于 Qwen/Qwen-Image-2512—660
mflux-community/ernie-image-turbo-mflux-q4MLX基于 baidu/ERNIE-Image-Turbo—646
mflux-community/krea-2-turbo-mflux-bf16MLX基于 krea/Krea-2-Turbo—642
ddalcu/Krea-2-Turbo-MLX-Serve-mixed-4-8MLX基于 krea/Krea-2-Turbo—627
ddalcu/Qwen-Image-2.1-Turbo-MLX-Serve-4bitMLX基于 Qwen/Qwen-Image-2.1-Turbo—622
mflux-community/flux-1-dev-fill-mflux-q6MLX基于 black-forest-labs/FLUX.1-Fill-dev—611
mflux-community/qwen-image-2-1-mflux-q6MLX基于 Qwen/Qwen-Image-2.1—609
SirSahOl/Qwen-Image-mlx-8bitMLX基于 Qwen/Qwen-Image—603
argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantizedMLX基于 stabilityai/stable-diffusion-3.5-large—580
apple/coreml-stable-diffusion-v1-5Core ML—63
apple/coreml-stable-diffusion-2-1-baseCore ML—47
devin-lai/Qwen-Image-2.1-CoremlCore ML基于 Qwen/Qwen-Image-2.1—38
mlboydaisuke/Hyper-SD-1step-CoreMLCore ML基于 ByteDance/Hyper-SD—37
digitalbrain79/sdxl-lightning-4step-coreml-6bits-compiledCore ML—29
c2p-cmd/StableDiffusionXLBase1.0-Palletized6.5Core ML—22
apple/coreml-stable-diffusion-2-1-base-palettizedCore ML—20
darkmaniac7/tokforge-ios-coreml-image-modelsCore ML基于 stabilityai/sdxl-turbo—20
apple/coreml-stable-diffusion-xl-baseCore ML—19
apple/coreml-stable-diffusion-v1-4Core ML—17
apple/coreml-stable-diffusion-2-baseCore ML—11
apple/coreml-stable-diffusion-v1-5-palettizedCore ML—11
apple/coreml-stable-diffusion-xl-base-with-refinerCore ML—11
apple/coreml-stable-diffusion-2-base-palettizedCore ML—9

如何阅读此表

  • 下载量是 Hugging Face 统计的最近 30 天数据，不是累计值。Hugging Face 按模型主文件的请求次数计数，所以一次安装可能被计入多次。
  • 大小是 Hugging Face 上该仓库当前全部文件的总和（1 GB = 1,000 MB）。有些仓库同时存放同一模型的多个版本，所以可能比你实际要下载的那一个更大；显示短横线表示还没有测量。
  • 热门是 Hugging Face 自己的近期关注度分数。名称下方的“基于”是该模型的来源模型。
  • 参数量是模型文件中的参数量。标有 * 的数值取自模型名称（例如 27B）：文件里没有参数量，或参数量小得多（打包的低比特文件会这样）。
  • 量化信息读取自模型的标签和名称，未对照文件核实。

使用之前

列表来自 Hugging Face 的公开 API，每天更新。我们不会逐个测试模型。在应用中使用前，请查看许可证并阅读模型卡片：有些许可证限制商业用途，“other”表示条款写在模型卡片里。名称或标签指向成人内容或去除安全限制的模型不会列出；需要登录 Hugging Face 才能下载的受限模型也不会列出。

其他任务