Upload folder using huggingface_hub

config.json

@@ -64,7 +64,7 @@
         "STATE": "MEAN_STD",
         "ACTION": "MEAN_STD"
     },
-    "vision_backbone": "vit_b_16",
+    "vision_backbone": "qwen3_vl_4b_instruct",
     "num_cameras": 3,
     "cameras_for_vision_state_concat": [
         "observation.images.front",
@@ -75,7 +75,7 @@
     "action_dim": 7,
     "d_model": 512,
     "nhead": 8,
-    "num_decoder_layers": 16,
+    "num_decoder_layers": 10,
     "dim_feedforward": 512,
     "diffusion_step_embed_dim": 256,
     "num_inference_steps": 20,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d79397240e1e242fe7197c97513be7937c3e8ba2ccb06d9fa3f8062694db18e7
-size 176268340
+oid sha256:d70e48c70855c2e78d9476dd34b0a5f30d0488413d39172093ccaafc39f03574
+size 113149260