Add gr00t_qwen3vl_0.6b_libero checkpoint

.gitattributes

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 pi05_libero/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 pi05_base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 pi0_base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gr00t_qwen3vl_0.6b_libero/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text

gr00t_qwen3vl_0.6b_libero/checkpoints/step-104160-epoch-24-loss=0.0550.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0c1a3af779899af469531fb8df0b4c5b7c1c6306a27e5ee00f68ac427cd3e6b
+size 25309396498

gr00t_qwen3vl_0.6b_libero/config.json

@@ -0,0 +1,462 @@
+{
+  "eval": {
+    "dataset": {
+      "transforms": [
+        {
+          "embodiment_id": 2,
+          "img_keys": [
+            "agentview_image",
+            "robot0_eye_in_hand_image"
+          ],
+          "type": "ProcessLiberoEvalInputs"
+        },
+        {
+          "type": "ConvertPILImageToNumpyArray"
+        },
+        {
+          "image_mean": [
+            0.48145466,
+            0.4578275,
+            0.40821073
+          ],
+          "image_std": [
+            0.26862954,
+            0.26130258,
+            0.27577711
+          ],
+          "img_key": "pixel_values",
+          "max_pixels": 1003520,
+          "merge_size": 2,
+          "min_pixels": 3136,
+          "patch_size": 16,
+          "temporal_patch_size": 2,
+          "to_tensor": true,
+          "type": "QWen2VLImageTransform"
+        },
+        {
+          "tokenizer": {
+            "model_path": "./checkpoints/LimX-Qwen3-VL-0.6B-stage1-20260223/",
+            "type": "PretrainedTokenizer"
+          },
+          "type": "LiberoPromptFromInputs"
+        },
+        {
+          "gripper_key": "robot0_gripper_qpos",
+          "norm_type": "mean_std",
+          "out_key": "states",
+          "pos_key": "robot0_eef_pos",
+          "quat_key": "robot0_eef_quat",
+          "state_dim": 64,
+          "type": "LiberoProprioFromInputs"
+        }
+      ],
+      "type": "LiberoParquetEvalDataset"
+    },
+    "denormalize_action": {
+      "norm_type": "mean_std",
+      "type": "DenormalizeLiberoAction"
+    },
+    "eval_chunk_size": 10,
+    "model_family": "pi0",
+    "num_steps_wait": 10,
+    "num_trials_per_task": 50,
+    "resize_size": 224,
+    "seed": 7,
+    "task_suite_name": "libero_10",
+    "type": "LiberoEvalRunner"
+  },
+  "inference_model": {
+    "freeze_projector": false,
+    "freeze_vlm_backbone": false,
+    "name_mapping": null,
+    "pretrained_name_or_path": null,
+    "type": "LlavaVLA",
+    "vla_head": {
+      "action_dim": 32,
+      "backbone_embedding_dim": 2048,
+      "diffusion_model_cfg": {
+        "attention_head_dim": 48,
+        "cross_attention_dim": 2048,
+        "dropout": 0.2,
+        "final_dropout": true,
+        "interleave_self_attention": true,
+        "norm_type": "ada_norm",
+        "num_attention_heads": 32,
+        "num_layers": 16,
+        "output_dim": 1024,
+        "positional_embeddings": null
+      },
+      "hidden_size": 1024,
+      "input_embedding_dim": 1536,
+      "num_heads": 4,
+      "num_inference_timesteps": 4,
+      "num_layers": 1,
+      "ori_action_dim": 7,
+      "state_dim": 64,
+      "traj_length": 10,
+      "type": "FlowMatchingHead",
+      "vl_self_attention_cfg": {
+        "attention_head_dim": 64,
+        "dropout": 0.2,
+        "final_dropout": true,
+        "num_attention_heads": 32,
+        "num_layers": 4,
+        "positional_embeddings": null
+      }
+    },
+    "vlm_backbone": {
+      "attn_implementation": "sdpa",
+      "projection_output_dim": 2048,
+      "projection_type": "linear",
+      "type": "Qwen3VL",
+      "use_projection": true,
+      "vlm_backbone_id": "qwen3_0.6b_vl_pt",
+      "vlm_config": null,
+      "vlm_path": "./check;points/LimX-Qwen3-VL-0.6B-stage1-20260223/"
+    }
+  },
+  "model": {
+    "freeze_projector": false,
+    "freeze_vlm_backbone": false,
+    "name_mapping": {
+      "vla_head": "action_head"
+    },
+    "pretrained_name_or_path": "./checkpoints/GR00T-N1.5-3B",
+    "strict_mapping": false,
+    "type": "LlavaVLA",
+    "vla_head": {
+      "action_dim": 32,
+      "backbone_embedding_dim": 2048,
+      "diffusion_model_cfg": {
+        "attention_head_dim": 48,
+        "cross_attention_dim": 2048,
+        "dropout": 0.2,
+        "final_dropout": true,
+        "interleave_self_attention": true,
+        "norm_type": "ada_norm",
+        "num_attention_heads": 32,
+        "num_layers": 16,
+        "output_dim": 1024,
+        "positional_embeddings": null
+      },
+      "hidden_size": 1024,
+      "input_embedding_dim": 1536,
+      "num_heads": 4,
+      "num_inference_timesteps": 4,
+      "num_layers": 1,
+      "ori_action_dim": 7,
+      "state_dim": 64,
+      "traj_length": 10,
+      "type": "FlowMatchingHead",
+      "vl_self_attention_cfg": {
+        "attention_head_dim": 64,
+        "dropout": 0.2,
+        "final_dropout": true,
+        "num_attention_heads": 32,
+        "num_layers": 4,
+        "positional_embeddings": null
+      }
+    },
+    "vlm_backbone": {
+      "attn_implementation": "sdpa",
+      "projection_output_dim": 2048,
+      "projection_type": "linear",
+      "type": "Qwen3VL",
+      "use_projection": true,
+      "vlm_backbone_id": "qwen3_0.6b_vl_pt",
+      "vlm_config": null,
+      "vlm_path": "./check;points/LimX-Qwen3-VL-0.6B-stage1-20260223/"
+    }
+  },
+  "runner": {
+    "change_key_name": false,
+    "collator": {
+      "keys": [
+        "states",
+        "observation.eepose",
+        "timestamp",
+        "images",
+        "img_masks",
+        "lang_tokens",
+        "lang_masks",
+        "actions",
+        "action_masks",
+        "embodiment_ids",
+        "image_grid_thw"
+      ],
+      "meta_keys": [
+        "task_description",
+        "prompt",
+        "info",
+        "stats"
+      ],
+      "type": "DictCollator"
+    },
+    "enable_gradient_checkpointing": false,
+    "enable_mixed_precision_training": true,
+    "learning_rate": 1.5e-05,
+    "lr_scheduler_type": "linear-warmup+cosine-decay",
+    "max_epochs": 24,
+    "max_grad_norm": 1.0,
+    "metric": {
+      "active_trackers": [
+        "jsonl",
+        "wandb"
+      ],
+      "grad_accumulation_steps": 1,
+      "run_dir": "work_dirs",
+      "type": "VLAMetric",
+      "window_size": 1
+    },
+    "mixed_precision_dtype": "bf16",
+    "sampler": null,
+    "sharding_strategy": "full-shard",
+    "tokenizer": {
+      "model_path": "./check;points/LimX-Qwen3-VL-0.6B-stage1-20260223/",
+      "type": "PretrainedTokenizer"
+    },
+    "type": "FSDPTrainRunner",
+    "warmup_ratio": 0.03,
+    "weight_decay": 0.0
+  },
+  "train_dataloader": {
+    "dataset": {
+      "dataset_statistics": {
+        "libero_10_no_noops": {
+          "action": {
+            "max": [
+              0.9375,
+              0.9375,
+              0.9375,
+              0.32892856001853943,
+              0.36964285373687744,
+              0.375,
+              1.0
+            ],
+            "mean": [
+              0.01905656634877842,
+              0.05672475971568838,
+              -0.056239289430234256,
+              0.004756678478841528,
+              0.002797492338491304,
+              -0.00714607048416358,
+              0.54599156235075
+            ],
+            "min": [
+              -0.9375,
+              -0.9375,
+              -0.9375,
+              -0.23642857372760773,
+              -0.3053571283817291,
+              -0.3642857074737549,
+              0.0
+            ],
+            "q01": [
+              -0.4997477764535965,
+              -0.6992653512084763,
+              -0.6543309163615124,
+              -0.07417070079989778,
+              -0.11898748445770971,
+              -0.15976085962510805,
+              0.0
+            ],
+            "q99": [
+              0.658747846713789,
+              0.7333480638990948,
+              0.768601965587579,
+              0.09784501244893279,
+              0.12943469061349036,
+              0.15137893471596325,
+              1.0
+            ],
+            "std": [
+              0.10588348353857541,
+              0.13552477199270377,
+              0.13886650724555177,
+              0.01433739270759898,
+              0.02038583948325967,
+              0.033299202425577934,
+              0.1881810653484855
+            ]
+          },
+          "proprio": {
+            "max": [
+              0.2103137969970703,
+              0.38887521624565125,
+              1.333192229270935,
+              3.7248642444610596,
+              3.5618896484375,
+              1.3863215446472168,
+              0.041575800627470016,
+              0.0013126095291227102
+            ],
+            "mean": [
+              -0.0419132679050224,
+              0.034591788297521735,
+              0.8265881844959498,
+              2.90259518190321,
+              -0.5570652600832564,
+              -0.16592166873533284,
+              0.02845031351083622,
+              -0.02880236273799356
+            ],
+            "min": [
+              -0.48278069496154785,
+              -0.3309336006641388,
+              0.44550687074661255,
+              1.1323540210723877,
+              -3.6312508583068848,
+              -1.842738389968872,
+              -0.005453015677630901,
+              -0.04112039878964424
+            ],
+            "q01": [
+              -0.1855636807291125,
+              -0.16145669766439186,
+              0.7064185725262808,
+              2.5678211534702324,
+              -1.2430377303522737,
+              -0.5195810482339626,
+              0.01022917473133343,
+              -0.03999379658232052
+            ],
+            "q99": [
+              0.05938728483051665,
+              0.2361478409238694,
+              0.9397258571145816,
+              3.2118708728143526,
+              0.49082919816100534,
+              0.2100883989120329,
+              0.040047131839991014,
+              -0.011104049991952391
+            ],
+            "std": [
+              0.03756502182067285,
+              0.05091765880150317,
+              0.09107525593038836,
+              0.12327524826514363,
+              0.4418352294043351,
+              0.12490994022681218,
+              0.004662133639412193,
+              0.00460807817987938
+            ]
+          },
+          "timestamp": {
+            "max": [
+              25.2
+            ],
+            "mean": [
+              7.007510548523206
+            ],
+            "min": [
+              0.0
+            ],
+            "q01": null,
+            "q99": null,
+            "std": [
+              4.457129586378845
+            ]
+          }
+        }
+      },
+      "datasets": {
+        "action_key": "action",
+        "action_window_size": 10,
+        "data_root_path": [
+          "datasets/libero_10_lerobot",
+          "datasets/libero_goal_no_noops_lerobotv2.1",
+          "datasets/libero_spatial_lerobot",
+          "datasets/libero_object_lerobot"
+        ],
+        "statistic_name": "libero_10_no_noops",
+        "transforms": [
+          {
+            "embodiment_id": 2,
+            "name_mappings": {
+              "actions": [
+                "actions"
+              ],
+              "observation.state": [
+                "states"
+              ]
+            },
+            "parquet_keys": [
+              "observation.state",
+              "timestamp",
+              "actions",
+              "info",
+              "stats",
+              "action_masks"
+            ],
+            "type": "ProcessParquetInputs",
+            "video_keys": [
+              "observation.images.image",
+              "observation.images.wrist_image"
+            ]
+          },
+          {
+            "type": "ParquetPrompter"
+          },
+          {
+            "tokenizer": {
+              "model_path": "./check;points/LimX-Qwen3-VL-0.6B-stage1-20260223/",
+              "type": "PretrainedTokenizer"
+            },
+            "type": "ProcessPrompts"
+          },
+          {
+            "height": 224,
+            "type": "ResizeImages",
+            "width": 224
+          },
+          {
+            "image_mean": [
+              0.48145466,
+              0.4578275,
+              0.40821073
+            ],
+            "image_std": [
+              0.26862954,
+              0.26130258,
+              0.27577711
+            ],
+            "max_pixels": 1003520,
+            "merge_size": 2,
+            "min_pixels": 3136,
+            "patch_size": 16,
+            "temporal_patch_size": 2,
+            "type": "QWen2VLImageTransform"
+          },
+          {
+            "action_dim": 32,
+            "action_key": "action",
+            "norm_type": "mean_std",
+            "state_dim": 64,
+            "state_key": "proprio",
+            "type": "NormalizeStatesAndActions"
+          }
+        ],
+        "type": "ParquetDataset",
+        "use_delta": false,
+        "window_start_idx": 0
+      },
+      "name_mappings": {
+        "action": [
+          "action"
+        ],
+        "observation.state": [
+          "proprio"
+        ]
+      },
+      "statistic_keys": [
+        "observation.state",
+        "timestamp",
+        "action"
+      ],
+      "statistic_name": "libero_10_no_noops",
+      "type": "DistributedRepeatingDataset"
+    },
+    "per_device_batch_size": 8,
+    "per_device_num_workers": 4
+  }
+}

gr00t_qwen3vl_0.6b_libero/config.yaml

@@ -0,0 +1,374 @@
+eval:
+  dataset:
+    transforms:
+    - embodiment_id: 2
+      img_keys:
+      - agentview_image
+      - robot0_eye_in_hand_image
+      type: ProcessLiberoEvalInputs
+    - type: ConvertPILImageToNumpyArray
+    - image_mean:
+      - 0.48145466
+      - 0.4578275
+      - 0.40821073
+      image_std:
+      - 0.26862954
+      - 0.26130258
+      - 0.27577711
+      img_key: pixel_values
+      max_pixels: 1003520
+      merge_size: 2
+      min_pixels: 3136
+      patch_size: 16
+      temporal_patch_size: 2
+      to_tensor: true
+      type: QWen2VLImageTransform
+    - tokenizer:
+        model_path: ./checkpoints/LimX-Qwen3-VL-0.6B-stage1-20260223/
+        type: PretrainedTokenizer
+      type: LiberoPromptFromInputs
+    - gripper_key: robot0_gripper_qpos
+      norm_type: mean_std
+      out_key: states
+      pos_key: robot0_eef_pos
+      quat_key: robot0_eef_quat
+      state_dim: 64
+      type: LiberoProprioFromInputs
+    type: LiberoParquetEvalDataset
+  denormalize_action:
+    norm_type: mean_std
+    type: DenormalizeLiberoAction
+  eval_chunk_size: 10
+  model_family: pi0
+  num_steps_wait: 10
+  num_trials_per_task: 50
+  resize_size: 224
+  seed: 7
+  task_suite_name: libero_10
+  type: LiberoEvalRunner
+inference_model:
+  freeze_projector: false
+  freeze_vlm_backbone: false
+  name_mapping: null
+  pretrained_name_or_path: null
+  type: LlavaVLA
+  vla_head:
+    action_dim: 32
+    backbone_embedding_dim: 2048
+    diffusion_model_cfg:
+      attention_head_dim: 48
+      cross_attention_dim: 2048
+      dropout: 0.2
+      final_dropout: true
+      interleave_self_attention: true
+      norm_type: ada_norm
+      num_attention_heads: 32
+      num_layers: 16
+      output_dim: 1024
+      positional_embeddings: null
+    hidden_size: 1024
+    input_embedding_dim: 1536
+    num_heads: 4
+    num_inference_timesteps: 4
+    num_layers: 1
+    ori_action_dim: 7
+    state_dim: 64
+    traj_length: 10
+    type: FlowMatchingHead
+    vl_self_attention_cfg:
+      attention_head_dim: 64
+      dropout: 0.2
+      final_dropout: true
+      num_attention_heads: 32
+      num_layers: 4
+      positional_embeddings: null
+  vlm_backbone:
+    attn_implementation: sdpa
+    projection_output_dim: 2048
+    projection_type: linear
+    type: Qwen3VL
+    use_projection: true
+    vlm_backbone_id: qwen3_0.6b_vl_pt
+    vlm_config: null
+    vlm_path: ./checkpoints/LimX-Qwen3-VL-0.6B-stage1-20260223/
+model:
+  freeze_projector: false
+  freeze_vlm_backbone: false
+  name_mapping:
+    vla_head: action_head
+  pretrained_name_or_path: ./checkpoints/GR00T-N1.5-3B
+  strict_mapping: false
+  type: LlavaVLA
+  vla_head:
+    action_dim: 32
+    backbone_embedding_dim: 2048
+    diffusion_model_cfg:
+      attention_head_dim: 48
+      cross_attention_dim: 2048
+      dropout: 0.2
+      final_dropout: true
+      interleave_self_attention: true
+      norm_type: ada_norm
+      num_attention_heads: 32
+      num_layers: 16
+      output_dim: 1024
+      positional_embeddings: null
+    hidden_size: 1024
+    input_embedding_dim: 1536
+    num_heads: 4
+    num_inference_timesteps: 4
+    num_layers: 1
+    ori_action_dim: 7
+    state_dim: 64
+    traj_length: 10
+    type: FlowMatchingHead
+    vl_self_attention_cfg:
+      attention_head_dim: 64
+      dropout: 0.2
+      final_dropout: true
+      num_attention_heads: 32
+      num_layers: 4
+      positional_embeddings: null
+  vlm_backbone:
+    attn_implementation: sdpa
+    projection_output_dim: 2048
+    projection_type: linear
+    type: Qwen3VL
+    use_projection: true
+    vlm_backbone_id: qwen3_0.6b_vl_pt
+    vlm_config: null
+    vlm_path: ./checkpoints/LimX-Qwen3-VL-0.6B-stage1-20260223/
+runner:
+  change_key_name: false
+  collator:
+    keys:
+    - states
+    - observation.eepose
+    - timestamp
+    - images
+    - img_masks
+    - lang_tokens
+    - lang_masks
+    - actions
+    - action_masks
+    - embodiment_ids
+    - image_grid_thw
+    meta_keys:
+    - task_description
+    - prompt
+    - info
+    - stats
+    type: DictCollator
+  enable_gradient_checkpointing: false
+  enable_mixed_precision_training: true
+  learning_rate: 1.5e-05
+  lr_scheduler_type: linear-warmup+cosine-decay
+  max_epochs: 24
+  max_grad_norm: 1.0
+  metric:
+    active_trackers:
+    - jsonl
+    - wandb
+    grad_accumulation_steps: 1
+    run_dir: work_dirs
+    type: VLAMetric
+    window_size: 1
+  mixed_precision_dtype: bf16
+  sampler: null
+  sharding_strategy: full-shard
+  tokenizer:
+    model_path: ./checkpoints/LimX-Qwen3-VL-0.6B-stage1-20260223/
+    type: PretrainedTokenizer
+  type: FSDPTrainRunner
+  warmup_ratio: 0.03
+  weight_decay: 0.0
+train_dataloader:
+  dataset:
+    dataset_statistics:
+      libero_10_no_noops:
+        action:
+          max:
+          - 0.9375
+          - 0.9375
+          - 0.9375
+          - 0.32892856001853943
+          - 0.36964285373687744
+          - 0.375
+          - 1.0
+          mean:
+          - 0.01905656634877842
+          - 0.05672475971568838
+          - -0.056239289430234256
+          - 0.004756678478841528
+          - 0.002797492338491304
+          - -0.00714607048416358
+          - 0.54599156235075
+          min:
+          - -0.9375
+          - -0.9375
+          - -0.9375
+          - -0.23642857372760773
+          - -0.3053571283817291
+          - -0.3642857074737549
+          - 0.0
+          q01:
+          - -0.4997477764535965
+          - -0.6992653512084763
+          - -0.6543309163615124
+          - -0.07417070079989778
+          - -0.11898748445770971
+          - -0.15976085962510805
+          - 0.0
+          q99:
+          - 0.658747846713789
+          - 0.7333480638990948
+          - 0.768601965587579
+          - 0.09784501244893279
+          - 0.12943469061349036
+          - 0.15137893471596325
+          - 1.0
+          std:
+          - 0.10588348353857541
+          - 0.13552477199270377
+          - 0.13886650724555177
+          - 0.01433739270759898
+          - 0.02038583948325967
+          - 0.033299202425577934
+          - 0.1881810653484855
+        proprio:
+          max:
+          - 0.2103137969970703
+          - 0.38887521624565125
+          - 1.333192229270935
+          - 3.7248642444610596
+          - 3.5618896484375
+          - 1.3863215446472168
+          - 0.041575800627470016
+          - 0.0013126095291227102
+          mean:
+          - -0.0419132679050224
+          - 0.034591788297521735
+          - 0.8265881844959498
+          - 2.90259518190321
+          - -0.5570652600832564
+          - -0.16592166873533284
+          - 0.02845031351083622
+          - -0.02880236273799356
+          min:
+          - -0.48278069496154785
+          - -0.3309336006641388
+          - 0.44550687074661255
+          - 1.1323540210723877
+          - -3.6312508583068848
+          - -1.842738389968872
+          - -0.005453015677630901
+          - -0.04112039878964424
+          q01:
+          - -0.1855636807291125
+          - -0.16145669766439186
+          - 0.7064185725262808
+          - 2.5678211534702324
+          - -1.2430377303522737
+          - -0.5195810482339626
+          - 0.01022917473133343
+          - -0.03999379658232052
+          q99:
+          - 0.05938728483051665
+          - 0.2361478409238694
+          - 0.9397258571145816
+          - 3.2118708728143526
+          - 0.49082919816100534
+          - 0.2100883989120329
+          - 0.040047131839991014
+          - -0.011104049991952391
+          std:
+          - 0.03756502182067285
+          - 0.05091765880150317
+          - 0.09107525593038836
+          - 0.12327524826514363
+          - 0.4418352294043351
+          - 0.12490994022681218
+          - 0.004662133639412193
+          - 0.00460807817987938
+        timestamp:
+          max:
+          - 25.2
+          mean:
+          - 7.007510548523206
+          min:
+          - 0.0
+          q01: null
+          q99: null
+          std:
+          - 4.457129586378845
+    datasets:
+      action_key: action
+      action_window_size: 10
+      data_root_path:
+      - datasets/libero_10_lerobot
+      - datasets/libero_goal_no_noops_lerobotv2.1
+      - datasets/libero_spatial_lerobot
+      - datasets/libero_object_lerobot
+      statistic_name: libero_10_no_noops
+      transforms:
+      - embodiment_id: 2
+        name_mappings:
+          actions:
+          - actions
+          observation.state:
+          - states
+        parquet_keys:
+        - observation.state
+        - timestamp
+        - actions
+        - info
+        - stats
+        - action_masks
+        type: ProcessParquetInputs
+        video_keys:
+        - observation.images.image
+        - observation.images.wrist_image
+      - type: ParquetPrompter
+      - tokenizer:
+          model_path: ./checkpoints/LimX-Qwen3-VL-0.6B-stage1-20260223/
+          type: PretrainedTokenizer
+        type: ProcessPrompts
+      - height: 224
+        type: ResizeImages
+        width: 224
+      - image_mean:
+        - 0.48145466
+        - 0.4578275
+        - 0.40821073
+        image_std:
+        - 0.26862954
+        - 0.26130258
+        - 0.27577711
+        max_pixels: 1003520
+        merge_size: 2
+        min_pixels: 3136
+        patch_size: 16
+        temporal_patch_size: 2
+        type: QWen2VLImageTransform
+      - action_dim: 32
+        action_key: action
+        norm_type: mean_std
+        state_dim: 64
+        state_key: proprio
+        type: NormalizeStatesAndActions
+      type: ParquetDataset
+      use_delta: false
+      window_start_idx: 0
+    name_mappings:
+      action:
+      - action
+      observation.state:
+      - proprio
+    statistic_keys:
+    - observation.state
+    - timestamp
+    - action
+    statistic_name: libero_10_no_noops
+    type: DistributedRepeatingDataset
+  per_device_batch_size: 8
+  per_device_num_workers: 4

gr00t_qwen3vl_0.6b_libero/dataset_statistics.json

@@ -0,0 +1,138 @@
+{
+  "libero_10_no_noops": {
+    "proprio": {
+      "mean": [
+        -0.0419132679050224,
+        0.034591788297521735,
+        0.8265881844959498,
+        2.90259518190321,
+        -0.5570652600832564,
+        -0.16592166873533284,
+        0.02845031351083622,
+        -0.02880236273799356
+      ],
+      "std": [
+        0.03756502182067285,
+        0.05091765880150317,
+        0.09107525593038836,
+        0.12327524826514363,
+        0.4418352294043351,
+        0.12490994022681218,
+        0.004662133639412193,
+        0.00460807817987938
+      ],
+      "min": [
+        -0.48278069496154785,
+        -0.3309336006641388,
+        0.44550687074661255,
+        1.1323540210723877,
+        -3.6312508583068848,
+        -1.842738389968872,
+        -0.005453015677630901,
+        -0.04112039878964424
+      ],
+      "max": [
+        0.2103137969970703,
+        0.38887521624565125,
+        1.333192229270935,
+        3.7248642444610596,
+        3.5618896484375,
+        1.3863215446472168,
+        0.041575800627470016,
+        0.0013126095291227102
+      ],
+      "q01": [
+        -0.1855636807291125,
+        -0.16145669766439186,
+        0.7064185725262808,
+        2.5678211534702324,
+        -1.2430377303522737,
+        -0.5195810482339626,
+        0.01022917473133343,
+        -0.03999379658232052
+      ],
+      "q99": [
+        0.05938728483051665,
+        0.2361478409238694,
+        0.9397258571145816,
+        3.2118708728143526,
+        0.49082919816100534,
+        0.2100883989120329,
+        0.040047131839991014,
+        -0.011104049991952391
+      ]
+    },
+    "timestamp": {
+      "mean": [
+        7.007510548523206
+      ],
+      "std": [
+        4.457129586378845
+      ],
+      "min": [
+        0.0
+      ],
+      "max": [
+        25.2
+      ],
+      "q01": null,
+      "q99": null
+    },
+    "action": {
+      "mean": [
+        0.01905656634877842,
+        0.05672475971568838,
+        -0.056239289430234256,
+        0.004756678478841528,
+        0.002797492338491304,
+        -0.00714607048416358,
+        0.54599156235075
+      ],
+      "std": [
+        0.10588348353857541,
+        0.13552477199270377,
+        0.13886650724555177,
+        0.01433739270759898,
+        0.02038583948325967,
+        0.033299202425577934,
+        0.1881810653484855
+      ],
+      "min": [
+        -0.9375,
+        -0.9375,
+        -0.9375,
+        -0.23642857372760773,
+        -0.3053571283817291,
+        -0.3642857074737549,
+        0.0
+      ],
+      "max": [
+        0.9375,
+        0.9375,
+        0.9375,
+        0.32892856001853943,
+        0.36964285373687744,
+        0.375,
+        1.0
+      ],
+      "q01": [
+        -0.4997477764535965,
+        -0.6992653512084763,
+        -0.6543309163615124,
+        -0.07417070079989778,
+        -0.11898748445770971,
+        -0.15976085962510805,
+        0.0
+      ],
+      "q99": [
+        0.658747846713789,
+        0.7333480638990948,
+        0.768601965587579,
+        0.09784501244893279,
+        0.12943469061349036,
+        0.15137893471596325,
+        1.0
+      ]
+    }
+  }
+}

gr00t_qwen3vl_0.6b_libero/tokenizer/added_tokens.json

@@ -0,0 +1,48 @@
+{
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<action_end>": 151675,
+  "<action_pad>": 151673,
+  "<action_start>": 151674,
+  "<action_video>": 151669,
+  "<box2d_end>": 151679,
+  "<box2d_start>": 151678,
+  "<future_video_pad>": 151670,
+  "<future_vision_end>": 151672,
+  "<future_vision_start>": 151671,
+  "<ignore_pad>": 151688,
+  "<point2d_end>": 151681,
+  "<point2d_start>": 151680,
+  "<ref_end>": 151683,
+  "<ref_keypoint_end>": 151685,
+  "<ref_keypoint_start>": 151684,
+  "<ref_start>": 151682,
+  "<think>": 151667,
+  "<think_end>": 151677,
+  "<think_start>": 151676,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<traj2d_end>": 151687,
+  "<traj2d_start>": 151686,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

gr00t_qwen3vl_0.6b_libero/tokenizer/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

gr00t_qwen3vl_0.6b_libero/tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b7a88bc82340dd4205b97b9c287df826cf386b31ac9ecd9e648073d940355e1
+size 11426476

gr00t_qwen3vl_0.6b_libero/tokenizer/tokenizer_config.json

@@ -0,0 +1,401 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151669": {
+      "content": "<action_video>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<future_video_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<future_vision_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<future_vision_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<action_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<action_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<action_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151676": {
+      "content": "<think_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151677": {
+      "content": "<think_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151678": {
+      "content": "<box2d_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151679": {
+      "content": "<box2d_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151680": {
+      "content": "<point2d_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151681": {
+      "content": "<point2d_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151682": {
+      "content": "<ref_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151683": {
+      "content": "<ref_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151684": {
+      "content": "<ref_keypoint_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151685": {
+      "content": "<ref_keypoint_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151686": {
+      "content": "<traj2d_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151687": {
+      "content": "<traj2d_end>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151688": {
+      "content": "<ignore_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen3VLAProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

gr00t_qwen3vl_0.6b_libero/vlm_backbone_config.json

@@ -0,0 +1,110 @@
+{
+  "action_pred_2d_waypoint_loss_weight": 1.0,
+  "action_pred_2d_waypoint_max_dim": 4,
+  "action_pred_action_head_max_num_embodiments": 64,
+  "action_pred_arm_dim": 12,
+  "action_pred_binary_flags_loss_weight": 1.0,
+  "action_pred_binary_flags_max_dim": 2,
+  "action_pred_diffusion_loss_weight": 1.0,
+  "action_pred_diffusion_transformer_gradient_weight": 1.0,
+  "action_pred_enable": false,
+  "action_pred_extra_camera_num": 2,
+  "action_pred_gripper_dim": 2,
+  "action_pred_gripper_loss_weight": 15.0,
+  "action_pred_hidden_size": 256,
+  "action_pred_loss_weight": 10.0,
+  "action_pred_max_cot_input_token": 50,
+  "action_pred_max_input_arm_state_dim": 32,
+  "action_pred_max_input_gripper_state_dim": 32,
+  "action_pred_model_type": "GRTransformer",
+  "action_pred_num_attention_heads": 8,
+  "action_pred_num_hidden_layers": 8,
+  "action_pred_objective": "continuous",
+  "action_pred_text_cot_gradient_weight": 1.0,
+  "action_pred_traj_length": 10,
+  "action_pred_tune_action_head_backbone": true,
+  "action_pred_tune_action_head_projector": true,
+  "action_pred_vision_cot_gradient_weight": 1.0,
+  "architectures": [
+    "Qwen3VLAForConditionalGeneration"
+  ],
+  "depth_pred_ds_ratio": 8,
+  "depth_pred_enable": false,
+  "depth_pred_head_type": "Unet",
+  "depth_pred_loss_type": "relative",
+  "depth_pred_loss_weight": 1.0,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "finetune_ignore_mismatched_sizes": false,
+  "image_token_id": 151655,
+  "mm_action_decoder_type": null,
+  "mm_spatial_pool_mode": "bilinear",
+  "mm_tunable_parts": "mm_vision_adapter,mm_language_model",
+  "model_type": "qwen3_vl",
+  "pad_token_id": 151643,
+  "pos_skipping_range": 4096,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": 151643,
+    "dtype": "bfloat16",
+    "eos_token_id": 151645,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 1024,
+    "initializer_range": 0.02,
+    "intermediate_size": 3072,
+    "max_position_embeddings": 262144,
+    "model_type": "qwen3_vl_text",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 28,
+    "num_key_value_heads": 8,
+    "pad_token_id": null,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        24,
+        20,
+        20
+      ],
+      "rope_theta": 5000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": true,
+    "use_cache": true,
+    "vocab_size": 151936
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.3.0.dev0",
+  "use_another_LLM_path": "",
+  "use_pos_skipping": false,
+  "video_pred_enable": false,
+  "video_pred_head_type": "mtp",
+  "video_pred_loss_weight": 0.5,
+  "video_pred_max_frame_length": 10,
+  "video_pred_max_mtp_depth": 10,
+  "video_pred_mtp_depth_num": 3,
+  "video_token_id": 151656,
+  "vision_config": {
+    "deepstack_visual_indexes": [
+      5,
+      11,
+      17
+    ],
+    "depth": 24,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1024,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "model_type": "qwen3_vl",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 1024,
+    "patch_size": 16,
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  }
+}