Skip to content

On-device models

Models you can run on iPhone, iPad and Mac, from Hugging Face

Core ML and MLX models published on Hugging Face, plus the models Apple publishes there, sorted by downloads in the last 30 days. Core ML models run on iPhone, iPad and Mac through Apple's Core ML framework; MLX models are made for Apple's MLX library, mostly on Apple silicon Macs (MLX Swift also runs on iPhone and iPad). Each name links to the model's page on Hugging Face.

1,009 Core ML models, 1,693 MLX models and 143 from Apple, updated daily from Hugging Face's public API. Last updated 2026-11-02 (UTC).

New this month

The most liked models created in the last 30 days, across Core ML, MLX and Apple's own.

See all new models

Showing 100 of 100.

ModelTaskSizeDownloads (30 days)
google-bert/bert-base-uncasedCore MLFill-mask3.5 GB29.2M
argmaxinc/whisperkit-coremlCore MLSpeech recognition30.8 GB27.3M
distilbert/distilgpt2Core MLText generation3.2 GB1.5M
thenlper/gte-smallCore MLSentence similarity803 MB1.4M
jinaai/jina-embeddings-v2-small-enCore MLEmbedding523 MB1.3M
jinaai/jina-embeddings-v2-base-enCore MLEmbedding2.2 GB357K
FluidInference/parakeet-tdt-0.6b-v3-coremlCore MLfrom nvidia/parakeet-tdt-0.6b-v3Speech recognition3.6 GB311K
argmaxinc/parakeetkit-proCore MLSpeech recognition3.5 GB200K
argmaxinc/qwenasrkit-proCore MLSpeech recognition3.9 GB101K
FluidInference/parakeet-ctc-110m-coremlCore MLfrom nvidia/parakeet-tdt_ctc-110mSpeech recognition106 MB91.3K
apple/coreml-depth-anything-v2-smallCore MLDepth estimation289 MB62.4K
FluidInference/silero-vad-coremlCore MLfrom onnx-community/silero-vadVoice activity detection8.4 MB54.5K
FluidInference/parakeet-realtime-eou-120m-coremlCore MLfrom nvidia/parakeet_realtime_eou_120m-v1Speech recognition1.3 GB54.1K
argmaxinc/speakerkit-coremlCore MLSpeech recognition33.1 MB52.5K
apple/mobilevit-xx-smallCore MLImage classification16.1 MB32.5K
tiiuae/falcon-7b-instructCore MLText generation56.6 GB32.1K
Barrymanalow/nb-whisper-coremlCore MLfrom NbAiLab/nb-whisper-smallSpeech recognition488 MB31.7K
aufklarer/Parakeet-EOU-120M-CoreML-INT8Core MLfrom nvidia/parakeet_realtime_eou_120m-v1Speech recognition118 MB29.2K
distilbert/distilbert-base-uncased-distilled-squadCore MLQuestion answering2 GB29.1K
FluidInference/speaker-diarization-coremlCore MLfrom pyannote/speaker-diarization-community-1Voice activity detection129 MB27.7K
aufklarer/Kokoro-82M-CoreMLCore MLfrom hexgrad/Kokoro-82MText to speech335 MB23.2K
Falconsai/text_summarizationCore MLSummarization1.6 GB22.5K
FluidInference/parakeet-ultra-coremlCore MLfrom moondream/parakeet-ultraSpeech recognition632 MB17.2K
FluidInference/kokoro-82m-coremlCore MLfrom hexgrad/Kokoro-82MText to speech5.6 GB16.1K
FluidInference/parakeet-tdt-0.6b-v2-coremlCore MLfrom nvidia/parakeet-tdt-0.6b-v2Speech recognition2.6 GB15.5K
argmaxinc/speakerkit-proCore MLSpeech recognition246 MB15.5K
argmaxinc/whisperkit-proCore MLSpeech recognition8.8 GB13.4K
altic-dev/nemotron-3-diarization-coremlCore MLfrom nvidia/Nemotron-3-DiarizationVoice activity detection199 MB11.7K
Gustavosta/MagicPrompt-Stable-DiffusionCore MLText generation2 GB8,628
argmaxinc/ttskit-coremlCore MLNo task listed3.8 GB7,988
apple/mobilevit-smallCore MLImage classification67.8 MB7,331
dearyoungjo/Whisper-Medicalv1CNB1Core MLNo task listed21.8 GB7,032
mattmireles/kokoro-coremlCore MLfrom hexgrad/Kokoro-82MText to speech3.4 GB6,909
BarathwajAnandan/cohere-transcribe-03-2026-CoreML-6bitCore MLfrom CohereLabs/cohere-transcribe-03-2026Speech recognition4.1 GB6,581
unum-cloud/uform3-image-text-multilingual-baseCore MLEmbedding2 GB6,177
desert-ant-labs/uhmCore MLAudio classification—5,609
aufklarer/Supertonic-3-CoreMLCore MLfrom Supertone/supertonic-3Text to speech—4,438
FluidInference/pocket-tts-coremlCore MLfrom kyutai/pocket-ttsText to speech—3,845
aufklarer/Omnilingual-ASR-CTC-300M-CoreML-INT8-10sCore MLfrom facebook/omniASR-CTC-300MSpeech recognition—3,824
apple/coreml-mobileclipCore MLNo task listed—3,814
TheStageAI/thewhisper-large-v3-turboCore MLfrom openai/whisper-large-v3-turboSpeech recognition—3,639
enterprise-explorers/Llama-2-7b-chat-coremlCore MLText generation—3,573
FluidInference/diar-streaming-sortformer-coremlCore MLfrom nvidia/diar_streaming_sortformer_4spk-v2.1Speech recognition—3,368
aufklarer/DeepFilterNet3-CoreMLCore MLAudio enhancement—3,195
FluidInference/parakeet-unified-en-0.6b-coremlCore MLfrom nvidia/parakeet-unified-en-0.6bSpeech recognition—3,178
AILogDev/Irodori-TTS-v4.1-Small-MF-CoreMLCore MLfrom Aratako/Irodori-TTS-v4.1-Small-MFText to speech—2,671
aufklarer/Silero-VAD-v5-CoreMLCore MLVoice activity detection—2,551
FluidInference/supertonic-3-coremlCore MLfrom Supertone/supertonic-3Text to speech—2,483
apple/coreml-sam2.1-tinyCore MLMask generation—2,393
Falconsai/medical_summarizationCore MLSummarization—2,260
FluidInference/embeddinggemma-2-coremlCore MLfrom google/embeddinggemma-2Embedding—2,131
ConwayResearch/Underdog-Bark-3.0-ProCore MLfrom nvidia/parakeet-tdt-0.6b-v3Speech recognition—2,079
gety-ai/gety-embed-v0Core MLfrom intfloat/multilingual-e5-smallSentence similarity—2,048
apple/coreml-sam2.1-smallCore MLMask generation—2,027
iliasaz/chatterbox-nano-coremlCore MLfrom ResembleAI/chatterbox-nanoText to speech—1,979
aufklarer/Parakeet-TDT-v3-CoreML-INT8-iOS-5sCore MLfrom nvidia/parakeet-tdt-0.6b-v2No task listed—1,814
BarathwajAnandan/nemotron-3.5-asr-streaming320-int8-CoreMLCore MLfrom nvidia/nemotron-3.5-asr-streaming-0.6bSpeech recognition—1,698
BarathwajAnandan/nemotron-3.5-asr-offline-6bit-CoreMLCore MLfrom nvidia/nemotron-3.5-asr-streaming-0.6bSpeech recognition—1,633
camstack/camstack-modelsCore MLEmbedding—1,626
vladimir94/dory-1.0-4b-ane-coremlCore MLfrom nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16Text generation—1,624
divergentlabs/masker-miniCore MLfrom divergentlabs/maskerNamed entity recognition—1,585
aufklarer/Qwen3-ASR-CoreMLCore MLfrom Qwen/Qwen3-ASR-0.6BSpeech recognition—1,582
aufklarer/Pyannote-Community-1-CoreMLCore MLfrom pyannote/speaker-diarization-community-1Audio classification—1,543
aufklarer/Nemotron-3.5-ASR-Streaming-0.6B-CoreML-INT8Core MLfrom nvidia/nemotron-3.5-asr-streaming-0.6bSpeech recognition—1,536
DictionLabs/whisperkit-coremlCore MLfrom openai/whisper-baseSpeech recognition—1,442
aufklarer/Silero-VAD-v6.2.1-CoreMLCore MLVoice activity detection—1,442
FluidInference/parakeet-tdt-ctc-110m-coremlCore MLfrom nvidia/parakeet-tdt_ctc-110mSpeech recognition—1,422
NateRunsA-LIST/alist-exercise-extractorCore MLNo task listed—1,416
aufklarer/Parakeet-TDT-v3-CoreML-INT8Core MLfrom nvidia/parakeet-tdt-0.6b-v2No task listed—1,194
seanll95/chatterbox-nano-coremlCore MLfrom ResembleAI/chatterbox-nanoText to speech—1,171
JumpDeer/Milan-s_Style_Transfer_MLModelsCore MLNo task listed—1,169
iliasaz/chatterbox-multilingual-coremlCore MLfrom ResembleAI/chatterboxText to speech—1,154
FluidInference/campplus-coremlCore MLAudio classification—1,060
apple/coreml-sam2.1-baseplusCore MLMask generation—1,054
FluidInference/localvqe-coremlCore MLfrom LocalAI-io/LocalVQENo task listed—1,052
FluidInference/unlimited-ocr-mlx-coremlCore MLfrom baidu/Unlimited-OCRVision language—1,047
aufklarer/Supertonic-3-CoreML-FP16Core MLfrom Supertone/supertonic-3Text to speech—1,023
mlboydaisuke/LaMa-CoreMLCore MLImage to image—1,012
scrypted/plugin-modelsCore MLNo task listed—991
ArgoSA/D-FINE-segCore MLObject detection—960
iliasaz/perth-coremlCore MLNo task listed—920
NateRunsA-LIST/alist-food-vision-so400m384Core MLNo task listed—901
iliasaz/chatterbox-turbo-coremlCore MLfrom ResembleAI/chatterbox-turboText to speech—898
FluidInference/d3-coremlCore MLfrom vllm-sr/d3-edgeZero-shot text classification—868
Yoshiyoshi1362/ovd-ios-modelsCore MLObject detection—820
vladimir94/lfm3-omni-2b-ane-coremlCore MLfrom LiquidAI/LFM3-Omni-2BText generation—799
apple/coreml-detr-semantic-segmentationCore MLImage segmentation—780
julien-c/hotdog-not-hotdogCore MLImage classification—749
usetape/speech-en-coremlCore MLNo task listed—724
yepher/screen-cut-pro-tts-coremlCore MLfrom ResembleAI/chatterboxText to speech—719
aufklarer/Whisper-Large-v3-Turbo-CoreMLCore MLfrom openai/whisper-large-v3-turboSpeech recognition—715
ales27pm/Dolphin3.0-CoreMLCore MLfrom dphn/Dolphin3.0-Llama3.2-3BText generation—686
wjbmattingly/kraken-yiddishCore MLNo task listed—679
unum-cloud/uform3-image-text-english-largeCore MLEmbedding—676
FluidInference/parakeet-redux-coremlCore MLfrom moondream/parakeet-reduxSpeech recognition—670
Norod78/distilgpt2-base-pretrained-heCore MLText generation—658
kimtrinh/resus-room-whisperCore MLfrom openai/whisper-large-v3-turboNo task listed—645
Writer/palmyra-smallCore MLText generation—638
apple/deeplabv3-mobilevit-xx-smallCore MLImage segmentation—604
TheStageAI/thewhisper-large-v3Core MLSpeech recognition—599

The 100 most downloaded are listed here. For the rest, browse them on Hugging Face. See more on Hugging Face

Browse by task

Pages for tasks with at least 5 models, across Core ML, MLX and Apple's own.

How to run these

  • Core ML: add the .mlpackage to your Xcode project, or download a compiled model when the app first needs it, and load it with the Core ML framework. Some repos ship their own Swift package; the model card says which.
  • MLX: on a Mac with Apple silicon, the mlx-lm and mlx-audio Python packages load these by name; inside an app, MLX Swift runs them on Mac, iPhone and iPad. Mind the memory: in our test a 4-bit 0.5B model peaked at about 0.36 GB on a Mac; larger models need far more.
  • Apple's own: some are Core ML files ready for an app; others are research weights for PyTorch and need converting with coremltools first.

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.