跳到正文

全部端侧模型

适用于 iPhone 和 Mac 的语音识别模型(Core ML 与 MLX)

该任务共有 253 个模型在 Hugging Face 上标为 Core ML 或 MLX,或由 Apple 发布:Core ML 224 个、MLX 29 个、其他 Apple 模型 0 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)
argmaxinc/whisperkit-coremlCore ML30.8 GB2733万
mlx-community/parakeet-tdt-0.6b-v3MLX基于 nvidia/parakeet-tdt-0.6b-v32.5 GB271万
mlx-community/parakeet-tdt-0.6b-v2MLX基于 nvidia/parakeet-tdt-0.6b-v22.5 GB80.7万
FluidInference/parakeet-tdt-0.6b-v3-coremlCore ML基于 nvidia/parakeet-tdt-0.6b-v33.6 GB31.1万
mlx-community/whisper-large-v3-turboMLX1.6 GB28.5万
argmaxinc/parakeetkit-proCore ML3.5 GB20万
mlx-community/whisper-small-mlxMLX481 MB11.7万
argmaxinc/qwenasrkit-proCore ML3.9 GB10.1万
FluidInference/parakeet-ctc-110m-coremlCore ML基于 nvidia/parakeet-tdt_ctc-110m106 MB9.1万
mlx-community/whisper-large-v3-mlxMLX3.1 GB7.2万
FluidInference/parakeet-realtime-eou-120m-coremlCore ML基于 nvidia/parakeet_realtime_eou_120m-v11.3 GB5.4万
argmaxinc/speakerkit-coremlCore ML33.1 MB5.2万
mlx-community/parakeet-tdt-1.1bMLX基于 nvidia/parakeet-tdt-1.1b4.3 GB3.3万
Barrymanalow/nb-whisper-coremlCore ML基于 NbAiLab/nb-whisper-small488 MB3.2万
aufklarer/Parakeet-EOU-120M-CoreML-INT8Core ML基于 nvidia/parakeet_realtime_eou_120m-v1118 MB2.9万
mlx-community/whisper-base-mlxMLX144 MB2.7万
mlx-community/whisper-medium-mlxMLX1.5 GB1.8万
FluidInference/parakeet-ultra-coremlCore ML基于 moondream/parakeet-ultra632 MB1.7万
FluidInference/parakeet-tdt-0.6b-v2-coremlCore ML基于 nvidia/parakeet-tdt-0.6b-v22.6 GB1.6万
argmaxinc/speakerkit-proCore ML246 MB1.5万
argmaxinc/whisperkit-proCore ML8.8 GB1.3万
mlx-community/whisper-large-v3-turbo-4bitMLX基于 openai/whisper-large-v3-turbo464 MB1.3万
FermionResearch/Phonon-2MLX基于 nvidia/parakeet-tdt-0.6b-v3164 MB1.1万
BarathwajAnandan/cohere-transcribe-03-2026-CoreML-6bitCore ML基于 CohereLabs/cohere-transcribe-03-20264.1 GB6,581
aufklarer/Omnilingual-ASR-CTC-300M-CoreML-INT8-10sCore ML基于 facebook/omniASR-CTC-300M—3,824
TheStageAI/thewhisper-large-v3-turboCore ML基于 openai/whisper-large-v3-turbo—3,639
FluidInference/diar-streaming-sortformer-coremlCore ML基于 nvidia/diar_streaming_sortformer_4spk-v2.1—3,368
mlx-community/parakeet-tdt_ctc-0.6b-jaMLX基于 nvidia/parakeet-tdt_ctc-0.6b-ja—3,206
FluidInference/parakeet-unified-en-0.6b-coremlCore ML基于 nvidia/parakeet-unified-en-0.6b—3,178
selcukkubur/parakeet-ultra-mlxMLX基于 moondream/parakeet-ultra—2,855
mlx-community/parakeet-tdt_ctc-110mMLX基于 nvidia/parakeet-tdt_ctc-110m—2,627
mlx-community/Voxtral-Mini-4B-Realtime-2602-4bitMLX基于 mistralai/Voxtral-Mini-4B-Realtime-2602—2,251
ConwayResearch/Underdog-Bark-3.0-ProCore ML基于 nvidia/parakeet-tdt-0.6b-v3—2,079
mlx-community/whisper-large-v3-turbo-8bitMLX基于 openai/whisper-large-v3-turbo—2,049
BarathwajAnandan/nemotron-3.5-asr-streaming320-int8-CoreMLCore ML基于 nvidia/nemotron-3.5-asr-streaming-0.6b—1,698
mlx-community/whisper-tiny-asr-fp16MLX—1,658
BarathwajAnandan/nemotron-3.5-asr-offline-6bit-CoreMLCore ML基于 nvidia/nemotron-3.5-asr-streaming-0.6b—1,633
aufklarer/Qwen3-ASR-CoreMLCore ML基于 Qwen/Qwen3-ASR-0.6B—1,582
aufklarer/Nemotron-3.5-ASR-Streaming-0.6B-CoreML-INT8Core ML基于 nvidia/nemotron-3.5-asr-streaming-0.6b—1,536
mlx-community/whisper-small-asr-fp16MLX—1,489
mlx-community/nemotron-3.5-asr-streaming-0.6bMLX基于 nvidia/nemotron-3.5-asr-streaming-0.6b—1,471
DictionLabs/whisperkit-coremlCore ML基于 openai/whisper-base—1,442
FluidInference/parakeet-tdt-ctc-110m-coremlCore ML基于 nvidia/parakeet-tdt_ctc-110m—1,422
FermionResearch/Phonon-2.5MLX基于 nvidia/parakeet-tdt-0.6b-v3—1,062
mlx-community/whisper-tiny.en-8bitMLX基于 openai/whisper-tiny.en—1,058
mlx-community/whisper-base-asr-fp16MLX—992
aufklarer/Qwen3-ASR-1.7B-MLX-4bitMLX基于 Qwen/Qwen3-ASR-1.7B—913
FermionResearch/Phonon-2.5-EnglishMLX基于 nvidia/parakeet-tdt-0.6b-v3—907
mlx-community/VibeVoice-ASR-4bitMLX—828
mlx-community/nemotron-3.5-asr-streaming-0.6b-8bitMLX基于 nvidia/nemotron-3.5-asr-streaming-0.6b—766
mlx-community/whisper-large-v3-asr-fp16MLX—728
aufklarer/Whisper-Large-v3-Turbo-CoreMLCore ML基于 openai/whisper-large-v3-turbo—715
mlx-community/whisper-large-v3-4bitMLX基于 openai/whisper-large-v3—678
FluidInference/parakeet-redux-coremlCore ML基于 moondream/parakeet-redux—670
ConwayResearch/Underdog-Bark-0.8B-1.0MLX—635
mlx-community/parakeet-ctc-1.1bMLX基于 nvidia/parakeet-ctc-1.1b—620
TheStageAI/thewhisper-large-v3Core ML—599
FluidInference/phonon-2-coremlCore ML基于 FermionResearch/Phonon-2—587
codeclawd/parakeet-tdt-0.6b-v3-coremlCore ML基于 nvidia/parakeet-tdt-0.6b-v3—539
FluidVoice-Labs/fluid-speech-mini-coremlCore ML基于 moondream/parakeet-ultra—517
nitinh/whisperkit-hinglish-coremlCore ML基于 Oriserve/Whisper-Hindi2Hinglish-Prime—516
baristiran/whisper-turbo-quranCore ML基于 openai/whisper-large-v3-turbo—493
FluidInference/sensevoice-small-coremlCore ML基于 FunAudioLLM/SenseVoiceSmall—393
FluidInference/parakeet-0.6b-ja-coremlCore ML基于 nvidia/parakeet-tdt_ctc-0.6b-ja—372
ConwayResearch/Underdog-Bark-3.0Core ML基于 nvidia/parakeet-tdt-0.6b-v3—358
FermionResearch/Phonon-2-CoreMLCore ML基于 FermionResearch/Phonon-2—347
yslinear/kotoba-whisper-v2.2-coremlCore ML基于 kotoba-tech/kotoba-whisper-v2.2—341
eaispart/whisperkit-coremlCore ML—316
markstrom/pianissimo-sv-coremlCore ML基于 KlangAI/pianissimo-sv—313
mweinbach1/parakeet-tdt-0.6b-v3-coremlCore ML基于 nvidia/parakeet-tdt-0.6b-v3—305
FluidInference/canary-1b-v2-coremlCore ML基于 nvidia/canary-1b-v2—289
eranshir/ivrit-ai-whisper-large-v3-turbo-coremlCore ML基于 ivrit-ai/whisper-large-v3-turbo—280
smkrv/gigaam-v3-e2e-rnnt-coremlCore ML基于 ai-sage/GigaAM-v3—256
VocaHQ/whisperkit-coremlCore ML基于 Oriserve/Whisper-Hindi2Hinglish-Apex—249
BuilderHelm/Parakeet-TDT-CTC-110M-CoreMLCore ML基于 nvidia/parakeet-tdt_ctc-110m—245
madjinn/kb-whisper-small-coremlCore ML基于 KBLab/kb-whisper-small—206
FluidInference/nemotron-3.5-transcribe-lite-coremlCore ML基于 nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B—187
DictionLabs/parakeet-tdt-0.6b-v3-coremlCore ML基于 nvidia/parakeet-tdt-0.6b-v3—185
Glimpse-Dictation/Parakeet-Ultra-coremlCore ML基于 moondream/parakeet-ultra—179
Selbyggen/nb-whisper-large-distil-turbo-coremlCore ML基于 NbAiLab/nb-whisper-large-distil-turbo-beta—176
BuilderHelm/Parakeet-CTC-110M-CoreMLCore ML基于 nvidia/parakeet-tdt_ctc-110m—167
ctyau/cohere-transcribe-coreml-int8Core ML—167
weiren119/Breeze-ASR-25-coreml-4bit-paletteCore ML基于 MediaTek-Research/Breeze-ASR-25—165
ctyau/cohere-transcribe-coreml-fp16Core ML基于 CohereLabs/cohere-transcribe-03-2026—161
FluidInference/qwen3-asr-0.6b-coremlCore ML基于 Qwen/Qwen3-ASR-0.6B—155
eranshir/ivrit-ai-whisper-large-v3-turbo-coreml-int8Core ML基于 ivrit-ai/whisper-large-v3-turbo—152
FluidInference/Nemotron-3.5-ASR-Streaming-Multilingual-0.6b-CoreMLCore ML基于 nvidia/nemotron-3.5-asr-streaming-0.6b—147
kramerthomas/roest-v3-whisper-1.5b-coremlCore ML基于 CoRal-project/roest-v3-whisper-1.5b—145
ConwayResearch/Underdog-Bark-3.0-PlusCore ML基于 nvidia/parakeet-tdt-0.6b-v3—140
FluidInference/nemotron-labs-unified-asr-0.6b-coremlCore ML基于 nvidia/nemotron-labs-unified-asr-0.6b—136
argmaxinc/whisperkit-coreml_01-30-24Core ML—135
weiren119/Breeze-ASR-26-coreml-4bit-paletteCore ML基于 MediaTek-Research/Breeze-ASR-26—135
spert/flix-swissgerman-whisperkitCore ML基于 Flix-AI/flix-swissgerman-full—124
gerardkieffer/diktatur-whisper-lb-coremlCore ML基于 unilux/whisper-base-v1-luxembourgish—118
ihnatov-yuri/parakeet-tdt-0.6b-v3-ukCore ML基于 nvidia/parakeet-tdt-0.6b-v3—117
Rybib/whisperkit-coremlCore ML—113
madjinn/kb-whisper-medium-coremlCore ML基于 KBLab/kb-whisper-medium—112
FluidInference/cohere-transcribe-03-2026-coremlCore ML基于 CohereLabs/cohere-transcribe-03-2026—110
jbilcke/whisper-medical-large-coremlCore ML基于 Na0s/Medical-Whisper-Large-v3—102
madjinn/whisper-medium-coremlCore ML基于 openai/whisper-medium—98

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。