Update README.md

README.md

@@ -4,7 +4,7 @@ license: apache-2.0
 language:
   - en
 base_model:
-  - SberRoboticsCenter/GreenVLA-5b-base-stride-1
+  - SberRoboticsCenter/GreenVLA-5b-base-stride-4
 pipeline_tag: robotics
 tags:
   - robotics
@@ -17,7 +17,7 @@ tags:
   - calvin
   - reinforcement-learning
 model-index:
-  - name: GreenVLA-5b-stride-1-R2-calvin
+  - name: GreenVLA-5b-stride-4-R2-calvin
     results:
       - task:
           type: robotics
@@ -30,7 +30,7 @@ model-index:
 
 <div align="center">
 
-# GreenVLA-5b-stride-1-R2-calvin
+# GreenVLA-5b-stride-4-R2-calvin
 
 ### RL-Aligned VLA for CALVIN
 
@@ -46,9 +46,9 @@ model-index:
 
 ## Overview
 
-**GreenVLA-5b-stride-1-R2-calvin** is the R2 (RL-aligned) checkpoint of the [Green-VLA](https://arxiv.org/abs/2602.00919) family, fine-tuned for the CALVIN benchmark environment.
+**GreenVLA-5b-stride-4-R2-calvin** is the R2 (RL-aligned) checkpoint of the [Green-VLA](https://arxiv.org/abs/2602.00919) family, fine-tuned for the CALVIN benchmark environment.
 
-Starting from [GreenVLA-5b-base-stride-1](https://huggingface.co/SberRoboticsCenter/GreenVLA-5b-base-stride-1), this model went through both R1 (supervised fine-tuning) and R2 (RL policy alignment) stages on CALVIN data.
+Starting from [GreenVLA-5b-base-stride-4](https://huggingface.co/SberRoboticsCenter/GreenVLA-5b-base-stride-4), this model went through both R1 (supervised fine-tuning) and R2 (RL policy alignment) stages on CALVIN data.
 
 ## Evaluation
 
@@ -62,7 +62,7 @@ Evaluated on the **CALVIN** benchmark:
 
 | | Details |
 |---|---|
-| **Base checkpoint** | [GreenVLA-5b-base-stride-1](https://huggingface.co/SberRoboticsCenter/GreenVLA-5b-base-stride-1) |
+| **Base checkpoint** | [GreenVLA-5b-base-stride-4](https://huggingface.co/SberRoboticsCenter/GreenVLA-5b-base-stride-4) |
 | **Stage** | R2 — RL policy alignment |
 | **Method** | Trajectory optimization (SFT + RL on collected trajectories) |
 | **Environment** | CALVIN |
@@ -91,7 +91,7 @@ from lerobot.common.utils.torch_observation import (
 
 # 1. Load policy and transforms.
 policy, input_transforms, output_transforms = load_pretrained_policy(
-    "SberRoboticsCenter/GreenVLA-5b-stride-1-R2-calvin",
+    "SberRoboticsCenter/GreenVLA-5b-stride-4-R2-calvin",
     data_config_name="calvin",
 )
 policy.to("cuda").eval()