Downloads · 30 days
15
6% of all-time downloads
pocket-tts-mlx
reiscook/pocket-tts-mlx
pocket-tts-mlx is a text-to-speech model from reiscook. Use it when you need text read aloud. It is set up for mlx. The card lists the license as cc-by-4.0.
MLX-converted weights for Kyutai's Pocket TTS model. Optimized for Apple Silicon inference via MLX (Python) and MLX-Swift.
All-time downloads
258
Public
Repo size
506 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors506 MB · 100%
At a glance
- Task
- Text-to-Speech
- Library
- mlx
- License
- cc-by-4.0
- Model type
- t5
- Access
- Public
- Created
- Feb 6, 2026
- Updated
- Feb 6, 2026
- SHA
- f3ff377e
Model card
From the Hugging Face model README
Pocket TTS - MLX Weights
MLX-converted weights for Kyutai's Pocket TTS model. Optimized for Apple Silicon inference via MLX (Python) and MLX-Swift.
Weight Variants
| Directory | Size | Description | RTF (M2 Air) |
|---|---|---|---|
bf16/ | 224MB | bfloat16 baseline | ~3x realtime |
int8/ | 148MB | 8-bit quantized FlowLM, bf16 Mimi | ~5x realtime |
int4/ | 107MB | 4-bit quantized FlowLM, bf16 Mimi | ~6x realtime |
All variants include FlowLM + Mimi decoder in a single unified mlx_model.safetensors file.
Voice Embeddings
8 pre-extracted voice embeddings from the Kyutai release:
voice/alba.safetensorsvoice/azelma.safetensorsvoice/cosette.safetensorsvoice/eponine.safetensorsvoice/fantine.safetensorsvoice/javert.safetensorsvoice/jean.safetensorsvoice/marius.safetensors
See GitHub repo for full source code. (coming soon)
Source
Converted from kyutai/pocket-tts official weights.
- Task
- Text-to-Speech
- Library
- mlx
- Type
- t5
- License
- cc-by-4.0
- Languages
- en
- Created
- Feb 6, 2026
- Updated
- Feb 6, 2026