Downloads · 30 days
6
14% of all-time downloads
smolvla-handshake-hand-landmarks-640x512
IntBot-ai/smolvla-handshake-hand-landmarks-640x512
smolvla-handshake-hand-landmarks-640x512 is a machine learning model from IntBot-ai. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This repository contains the IntBot SmolVLA checkpoint trained for the handshake task with:
All-time downloads
43
Public
Parameters
450M
907 MB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors907 MB · 100%
Parameter types
How the weights are stored.
BF16447M · 99%
Model card
From the Hugging Face model README
IntBot SmolVLA Handshake With Hand Landmarks 640x512
This repository contains the IntBot SmolVLA checkpoint trained for the handshake task with:
- two RGB image inputs at 640x512
- robot state
- hand landmark features
Training summary:
- policy:
intbot_smolvla - dataset:
IntBot-ai/handshake-dev-merged-260325-640x512-hand-landmarks-raw-image2 - checkpoint:
40000 - job name:
smolvla-handshake-hand-landmarks-640x512-train95-eval05-imgaug-strong
Local source checkpoint:
outputs/train/intbot_smolvla/smolvla-handshake-hand-landmarks-640x512-train95-eval05-imgaug-strong/checkpoints/040000/pretrained_model
Expected inputs:
observation.images.image1:(3, 512, 640)observation.images.image2:(3, 512, 640)observation.state:(38,)observation.hand_landmarks:(130,)
This is the exported pretrained_model directory from the final 40k-step checkpoint.
- Created
- Apr 8, 2026
- Updated
- Apr 8, 2026