Parakeet Tdt 0.6b v3 Coreml
Core ML build of nvidia/parakeet-tdt-0.6b-v3 (FastConformer encoder + TDT decoder, 0.6B parameters, 25 European languages including Russian and Ukrainian, punctuation and capitalization built in).
Model source
- Kind
- Other
- File size
- 243 B–425 MB
- Base model
- nvidia/parakeet-tdt-0.6b-v3
- Version
- vRevision 738ae6a
- Updated
- Oct 31, 2026
- Downloads
- 42
- Source
- Hugging Face
Source description
parakeet-tdt-0.6b-v3 - Core ML
Core ML build of nvidia/parakeet-tdt-0.6b-v3 (FastConformer encoder + TDT decoder, 0.6B parameters, 25 European languages including Russian and Ukrainian, punctuation and capitalization built in).
Mirrored from FluidInference/parakeet-tdt-0.6b-v3-coreml for the Letopis on-device transcription pipeline; the runtime is the FluidAudio Swift package (Apache-2.0). This mirror carries the single runtime file set the app downloads (fp16/int8 inference path).
Files
Preprocessor.mlmodelc,Encoder.mlmodelc,Decoder.mlmodelc,JointDecisionv3.mlmodelcconfig.json,parakeet_vocab.json,parakeet_v3_vocab.json
Credits and license
- Model: NVIDIA parakeet-tdt-0.6b-v3, CC-BY-4.0.
- Core ML conversion: FluidInference.
- No changes to the weights; this is a byte-identical mirror of the runtime files.
Sources
1 source
Sources
1 sourceVerified Oct 31
Download files
Repository files
File inventory only; completeness is unverified.
- Decoder.mlmodelc/analytics/coremldata.bin
- Decoder.mlmodelc/coremldata.bin
- Decoder.mlmodelc/metadata.json
- Decoder.mlmodelc/weights/weight.bin
- Encoder.mlmodelc/analytics/coremldata.bin
- Encoder.mlmodelc/coremldata.bin
- Encoder.mlmodelc/metadata.json
- Encoder.mlmodelc/weights/weight.bin
- JointDecisionv3.mlmodelc/analytics/coremldata.bin
- JointDecisionv3.mlmodelc/coremldata.bin
- JointDecisionv3.mlmodelc/metadata.json
- JointDecisionv3.mlmodelc/weights/weight.bin
- Preprocessor.mlmodelc/analytics/coremldata.bin
- Preprocessor.mlmodelc/coremldata.bin
- Preprocessor.mlmodelc/metadata.json
- Preprocessor.mlmodelc/weights/weight.bin
- config.json
- parakeet_v3_vocab.json
- parakeet_vocab.json
Model artifacts
9 artifacts
Model artifacts
9 artifactsDecoder.mlmodelc/analytics/coremldata.bin
bin · 243 B · SHA-256 4238c4e81ecd…a350 · Hugging Face
Download fileDecoder.mlmodelc/weights/weight.bin
bin · 22.5 MB · SHA-256 48adf0f0d47c…5d41 · Hugging Face
Download fileEncoder.mlmodelc/analytics/coremldata.bin
bin · 243 B · SHA-256 42e638870d73…a105 · Hugging Face
Download fileEncoder.mlmodelc/weights/weight.bin
bin · 425 MB · SHA-256 e2020f323703…f421 · Hugging Face
Download fileJointDecisionv3.mlmodelc/analytics/coremldata.bin
bin · 243 B · SHA-256 26def4bf73dd…7b6c · Hugging Face
Download fileJointDecisionv3.mlmodelc/coremldata.bin
bin · 521 B · SHA-256 f5fc08b74140…8342 · Hugging Face
Download fileJointDecisionv3.mlmodelc/weights/weight.bin
bin · 12.1 MB · SHA-256 4e0e63d84003…db3e · Hugging Face
Download fileSource excerpts
3 excerpts
Source excerpts
3 excerpts--- license: cc-by-4.0 language: - en - ru - uk - de - fr - es - it - pl - nl - pt tags: - automatic-speech-recognition - coreml - parakeet - fastconformer - tdt base_model: nvidia/parakeet-tdt-0.6b-v3 --- # parakeet-tdt-0.6b-v3 - Core ML Core ML build of [nvidia/parakeet-tdt-0.6b-v3](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3) (FastConformer encoder + TDT decoder, 0.6B parameters, 25 European languages including Russian and Ukrainian, punctuation and capitalization built in). Mirrored from [FluidInference/parakeet-tdt-0.6b-v3-coreml](https://huggingface.co/FluidInference/parakeet-tdt-0.6b-v3-coreml) for the [Letopis](https://letopisapp.com) on-device transcription pipeline; the runtime is the [FluidAudio](https://github.com/FluidInference/FluidAudio) Swift package (Apache-2.0). This mirror carries the single runtime file set the app downloads (fp16/int8 inference path). ## Files - `Preprocessor.mlmodelc`, `Encoder.mlmodelc`, `Decoder.mlmodelc`, `JointDecisionv3.mlmodelc` - `config.json`, `parakeet_vocab.json`, `parakeet_v3_vocab.json` ## Credits and license - Model: NVIDIA parakeet-tdt-0.6b-v3, [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/). - Core ML conversion: FluidInference. - No changes to the weights; this is a byte-identical mirror of the runtime files.
Source context: 42 downloads · 1 likes · Pipeline automatic-speech-recognition · Repo smkrv/parakeet-tdt-0.6b-v3-coreml
Source context: 17 downloads · 1 likes · Pipeline automatic-speech-recognition · Repo smkrv/parakeet-tdt-0.6b-v3-coreml
