pipecat-ai · First MLX build 18 Feb 2026
smart-turn-v3
MLX
Audio bsd-2-clause
A 8M parameter voice activity detection model from pipecat-ai, with one MLX build.
- Parameters
- 8M From Hugging Face
- Quantisation
- Unstated smart-turn-v3
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 32 MB Weights, estimated
- Downloads
- 734 This build, last 30 days
- Task
- Voice activity detection Audio
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at Unstated
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/smart-turn-v3MLX builds
1 on mlx-community| smart-turn-v3 | Unstated | 32 MB | 3 GB | 734 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.