Skip to content
ModelListedExternal
Download

Vitpose Base Simple Mlx

ViTPose (vitpose-base-simple) converted to MLX for on-device human pose estimation on Apple Silicon. Weights are float16.

Sourcehuggingface.co/nazarkozak/vitpose-base-simple-mlx/tree/8f0b20af3958dda668a60ca536e3e4c6b651fe3f
KindOtherBase modelusyd-community/vitpose-base-simpleVersionRevision 8f0b20aBgrade

Model source

Kind
Other
File size
164 MB
Base model
usyd-community/vitpose-base-simple
Version
vRevision 8f0b20a
Updated
Oct 10, 2026
Downloads
14
Source
Hugging Face

Source description

ViTPose base-simple — MLX

ViTPose (vitpose-base-simple) converted to MLX for on-device human pose estimation on Apple Silicon. Weights are float16.

Built for MLXPose — a native MLX Swift ViTPose implementation. The Swift forward pass is numerically verified against the Hugging Face reference (heatmaps max|Δ|=1.5e-6, decoded keypoints max 3e-5 px).

  • Backbone: plain ViT-base (12 layers, dim 768), patch 16, input 256×192.
  • Head: simple decoder → 17 COCO keypoint heatmaps (64×48).
  • Conversion: convert_vitpose_to_mlx.py.

Files

  • weights.safetensors — MLX float16 weights.
  • config.json — original ViTPose config.

License

Apache-2.0. Pretrained weights derive from COCO/MPII training data — review dataset terms for your use case.

Sources

1 source
Hugging FacePrimaryVersion 8f0b20af3958dda668a60ca536e3e4c6b651fe3f
huggingface.co/nazarkozak/vitpose-base-simple-mlx/tree/8f0b20af3958dda668a60ca536e3e4c6b651fe3f

Verified Oct 10

Model artifacts

1 artifact

weights.safetensors

safetensors · 164 MB · SHA-256 7bf89e96616c…735b · Hugging Face

Download file

Source excerpts

3 excerpts
ReadmeAug 1huggingface.co/nazarkozak/vitpose-base-simple-mlx

--- license: apache-2.0 tags: - pose-estimation - vitpose - mlx - mlx-swift - on-device - apple-silicon - keypoint-detection library_name: mlx base_model: usyd-community/vitpose-base-simple --- # ViTPose base-simple — MLX ViTPose (`vitpose-base-simple`) converted to **MLX** for on-device human pose estimation on Apple Silicon. Weights are float16. Built for **[MLXPose](https://github.com/NazarKozak/MLXPose)** — a native MLX Swift ViTPose implementation. The Swift forward pass is numerically verified against the Hugging Face reference (heatmaps `max|Δ|=1.5e-6`, decoded keypoints `max 3e-5 px`). - Backbone: plain ViT-base (12 layers, dim 768), patch 16, input 256×192. - Head: simple decoder → 17 COCO keypoint heatmaps (64×48). - Conversion: [`convert_vitpose_to_mlx.py`](https://github.com/NazarKozak/MLXPose/blob/main/scripts/convert_vitpose_to_mlx.py). ## Files - `weights.safetensors` — MLX float16 weights. - `config.json` — original ViTPose config. ## License Apache-2.0. Pretrained weights derive from COCO/MPII training data — review dataset terms for your use case.

NoteOct 10huggingface.co/nazarkozak/vitpose-base-simple-mlx

Source context: 14 downloads · 0 likes · Pipeline keypoint-detection · Library mlx · Repo nazarkozak/vitpose-base-simple-mlx

NoteAug 1huggingface.co/nazarkozak/vitpose-base-simple-mlx

Source context: 11 downloads · 0 likes · Pipeline keypoint-detection · Library mlx · Repo nazarkozak/vitpose-base-simple-mlx