yinchenghust
/

SimpleMemVLA-RoboDojo-Sim

vision-language-action

Model card Files Files and versions

SimpleMemVLA RoboDojo-Sim

SimpleMemVLA checkpoint for the RoboDojo simulation benchmark (dual-arm ARX X5, action_type=joint).

Paper: SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models
Code: OpenBMB/SimpleMemVLA
Evaluation: policy/SimpleMemVLA in XPolicyLab

Download

modelscope download keithyc/SimpleMemVLA-RoboDojo-Sim --local_dir checkpoints/SimpleMemVLA-RoboDojo-Sim

Downloads last month: 15

Safetensors

Model size

6B params

Tensor type

BF16

·

Video Preview

loading

Model tree for yinchenghust/SimpleMemVLA-RoboDojo-Sim

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

(959)

this model

Collection including yinchenghust/SimpleMemVLA-RoboDojo-Sim

SimpleMemVLA

A Simple but Effective Native-Video Memory for Vision-Language-Action Models • 12 items • Updated 3 days ago • 1

Paper for yinchenghust/SimpleMemVLA-RoboDojo-Sim

SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models

Paper • 2609.05533 • Published Sep 2 • 15