smolvla_black_cube_fullft

Mehul2112/smolvla_black_cube_fullft

smolvla_black_cube_fullft is a robotics model from Mehul2112. Use it for the robotics task on the model card, and read the license before you ship it in a product. It is set up for lerobot.

Full fine-tune of lerobot/smolvlabase on Mehul2112/so100blackcubemerged (SO-100, single front camera, 130 episodes).

Downloads · 30 days

11% of all-time downloads

All-time downloads

Public

Parameters

450M

1.8 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors907 MB · 100%

Parameter types

How the weights are stored.

BF16447M · 99%

Base models

lerobot/smolvla_base

Model card

From the Hugging Face model README

SmolVLA — so100_black_cube (full fine-tune)

Full fine-tune of lerobot/smolvla_base on Mehul2112/so100_black_cube_merged (SO-100, single front camera, 130 episodes).

Recipe

Mode: full fine-tune — VLM backbone unfrozen (train_expert_only=false, freeze_vision_encoder=false).
Learnable params: ~403M of ~450M (vision encoder + language model + action expert).
Batch size: 32 · AMP: bf16 · Optimizer LR: ~1e-4 with decay.
Hardware: single RTX 4090 (24 GB).

Checkpoint

This repo holds the step-14000 checkpoint.

Step	Train loss	Eval loss (15% held-out)
2000	0.186	0.314 ← eval minimum
4000	0.137	0.376
6000	0.105	0.414
8000	0.085	0.476
10000	0.069	0.565
12000	0.055	0.643
14000	0.048	0.723
16000	0.047	0.759
18000	~0.047	0.784
20000	~0.047	0.787

Note: Eval loss reached its minimum at step 2000 and rose monotonically afterward (overfitting on a small 110-episode training split). This step-14000 checkpoint has a very low training loss (0.048) but is well past the generalization optimum. For best real-robot generalization, the step-2000 checkpoint is preferable.

Usage

from lerobot.policies.smolvla.modeling_smolvla import SmolVLAPolicy
policy = SmolVLAPolicy.from_pretrained("Mehul2112/smolvla_black_cube_fullft")

Task: Robotics
Library: lerobot
Created: Aug 2, 2026
Updated: Aug 2, 2026