Upload GR00T Eagle 3B RoboCasa GR1 24x30 fine-tuned checkpoint

.gitattributes

@@ -40,3 +40,4 @@ gr00t_qwen3vl_0.6b_libero/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs
 pi05_paligemma_libero_10_full_finetune_bs64/pi05_paligemma_libero_10_full_finetune_2026_05_15_09_15_10.jsonl filter=lfs diff=lfs merge=lfs -text
 pi0_paligemma_libero_10_full_finetune_bs64/pi0_paligemma_libero_10_full_finetune_2026_05_25_21_16_07.jsonl filter=lfs diff=lfs merge=lfs -text
 openvla_dino_siglip_llama2_libero_10_full_finetune_bs64/openvla_dino_siglip_llama2_libero_10_full_finetune_2026_05_27_09_35_11.jsonl filter=lfs diff=lfs merge=lfs -text
+gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug_2026_05_26_16_59_09.jsonl filter=lfs diff=lfs merge=lfs -text

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/checkpoints/step-010000-epoch-14-loss=0.0061.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5989eea1bba8bfdd77f1afc7f5bf60118fba126fe75cc384aec82fc39b86b85e
+size 5448457056

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/config.json

@@ -0,0 +1,363 @@
+{
+  "_OFFICIAL_GR1_STATS_PATH": "work_dirs/official_groot_gr1_dataset_statistics.json",
+  "_STAT": "robocasa_gr1_24tasks_30ep",
+  "eval": {
+    "dataset": {
+      "transforms": [
+        {
+          "center_crop_scale": 0.95,
+          "embodiment_id": 24,
+          "img_key": "video.ego_view_bg_crop_pad_res256_freq20",
+          "normalize": false,
+          "resize_size": 224,
+          "type": "ProcessRobocasaEvalInputs"
+        },
+        {
+          "type": "RobocasaGR1N15Bridge"
+        },
+        {
+          "action_key": "action",
+          "norm_type": "min_max",
+          "normalize_states": false,
+          "state_dim": 64,
+          "state_key": "proprio",
+          "type": "NormalizeStatesAndActions"
+        },
+        {
+          "max_len": 900,
+          "num_images": 1,
+          "return_text": true,
+          "tokenizer": {
+            "model_path": "fluxvla/models/third_party_models/eagle2_hg_model",
+            "type": "PretrainedTokenizer"
+          },
+          "type": "ProcessPromptsWithImage"
+        },
+        {
+          "means": [
+            [
+              127.5,
+              127.5,
+              127.5
+            ]
+          ],
+          "stds": [
+            [
+              127.5,
+              127.5,
+              127.5
+            ]
+          ],
+          "type": "NormalizeImages"
+        }
+      ],
+      "type": "RobocasaEvalDataset",
+      "unnorm_key": "robocasa_gr1_24tasks_30ep"
+    },
+    "denormalize_action": {
+      "action_dim": 29,
+      "clip_actions": false,
+      "norm_type": "min_max",
+      "stats_order": "fluxvla",
+      "type": "DenormalizeRobocasaAction"
+    },
+    "eval_chunk_size": 16,
+    "max_episode_steps": 720,
+    "model_family": "groot",
+    "num_trials_per_task": 20,
+    "seed": 7,
+    "task_list": [
+      "gr1_unified/PnPBottleToCabinetClose_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PnPCanToDrawerClose_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PnPCupToDrawerClose_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PnPMilkToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PnPPotatoToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PnPWineToCabinetClose_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromCuttingboardToBasketSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromCuttingboardToPanSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromCuttingboardToPotSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlacematToBasketSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlacematToBowlSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlacematToPlateSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlacematToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlateToBowlSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlateToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlateToPanSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromPlateToPlateSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromTrayToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromTrayToPlateSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromTrayToPotSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromTrayToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env",
+      "gr1_unified/PosttrainPnPNovelFromTrayToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env"
+    ],
+    "type": "RobocasaEvalRunner",
+    "unnorm_key": "robocasa_gr1_24tasks_30ep"
+  },
+  "inference_model": {
+    "pretrained_name_or_path": "./checkpoints/GR00T-N1.5-3B",
+    "type": "LlavaVLA",
+    "vla_head": {
+      "action_dim": 32,
+      "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,
+      "max_input_seq_len": 900,
+      "num_heads": 4,
+      "num_inference_timesteps": 4,
+      "num_layers": 1,
+      "num_steps": 16,
+      "ori_action_dim": 29,
+      "state_dim": 64,
+      "traj_length": 16,
+      "type": "FlowMatchingInferenceHead"
+    },
+    "vlm_backbone": {
+      "type": "EagleInferenceBackbone",
+      "vlm_config": {
+        "max_input_seq_len": 900
+      },
+      "vlm_path": "fluxvla/models/third_party_models/eagle2_hg_model"
+    }
+  },
+  "model": {
+    "freeze_projector": false,
+    "freeze_vlm_backbone": false,
+    "name_mapping": {
+      "vla_head": "action_head",
+      "vlm_backbone.vlm": "backbone.eagle_model"
+    },
+    "pretrained_name_or_path": "./checkpoints/GR00T-N1.5-3B",
+    "type": "LlavaVLA",
+    "vla_head": {
+      "action_dim": 32,
+      "hidden_size": 1024,
+      "input_embedding_dim": 1536,
+      "num_heads": 4,
+      "num_inference_timesteps": 4,
+      "num_layers": 1,
+      "num_steps": 16,
+      "ori_action_dim": 29,
+      "state_dim": 64,
+      "traj_length": 16,
+      "type": "FlowMatchingHead"
+    },
+    "vlm_backbone": {
+      "dtype": "bf16",
+      "tune_llm": false,
+      "tune_visual": true,
+      "type": "EagleBackbone",
+      "vlm_config": {
+        "max_input_seq_len": 900
+      },
+      "vlm_path": "fluxvla/models/third_party_models/eagle2_hg_model"
+    }
+  },
+  "runner": {
+    "change_key_name": false,
+    "collator": {
+      "keys": [
+        "states",
+        "observation.eepose",
+        "timestamp",
+        "images",
+        "img_masks",
+        "lang_tokens",
+        "lang_masks",
+        "actions",
+        "action_masks",
+        "embodiment_ids"
+      ],
+      "meta_keys": [
+        "task_description",
+        "prompt",
+        "info",
+        "stats"
+      ],
+      "type": "DictCollator"
+    },
+    "enable_gradient_checkpointing": false,
+    "enable_mixed_precision_training": true,
+    "learning_rate": 6e-05,
+    "lr_scheduler_type": "linear-warmup+cosine-decay",
+    "max_epochs": null,
+    "max_grad_norm": 1.0,
+    "max_keep_ckpts": 20,
+    "max_steps": 50000,
+    "metric": {
+      "active_trackers": [
+        "jsonl",
+        "wandb"
+      ],
+      "grad_accumulation_steps": 1,
+      "run_dir": "work_dirs",
+      "type": "VLAMetric",
+      "window_size": 1
+    },
+    "mixed_precision_dtype": "bf16",
+    "sampler": null,
+    "save_epoch_interval": 1,
+    "save_full_model": true,
+    "save_iter_interval": 5000,
+    "sharding_strategy": "no-shard",
+    "tokenizer": {
+      "model_path": "fluxvla/models/third_party_models/eagle2_hg_model",
+      "type": "PretrainedTokenizer"
+    },
+    "type": "FSDPTrainRunner",
+    "warmup_ratio": 0.05,
+    "weight_decay": 1e-05
+  },
+  "train_dataloader": {
+    "dataset": {
+      "dataset_statistics_path": "work_dirs/official_groot_gr1_dataset_statistics.json",
+      "datasets": [
+        {
+          "action_key": "action",
+          "action_window_size": 16,
+          "data_root_path": [
+            "./datasets/robocasa_fluxvla/PnPBottleToCabinetClose",
+            "./datasets/robocasa_fluxvla/PnPCanToDrawerClose",
+            "./datasets/robocasa_fluxvla/PnPCupToDrawerClose",
+            "./datasets/robocasa_fluxvla/PnPMilkToMicrowaveClose",
+            "./datasets/robocasa_fluxvla/PnPPotatoToMicrowaveClose",
+            "./datasets/robocasa_fluxvla/PnPWineToCabinetClose",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToBasketSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPanSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPotSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBasketSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBowlSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToPlateSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToTieredshelfSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToBowlSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToCardboardboxSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPanSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPlateSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToCardboardboxSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPlateSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPotSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredbasketSplitA",
+            "./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredshelfSplitA"
+          ],
+          "statistic_name": "robocasa_gr1_24tasks_30ep",
+          "transforms": [
+            {
+              "embodiment_id": 24,
+              "name_mappings": {
+                "actions": [
+                  "actions"
+                ],
+                "observation.state": [
+                  "states"
+                ]
+              },
+              "parquet_keys": [
+                "observation.state",
+                "timestamp",
+                "actions",
+                "info",
+                "stats",
+                "action_masks"
+              ],
+              "type": "ProcessParquetInputs",
+              "video_keys": [
+                "observation.images.ego_view"
+              ]
+            },
+            {
+              "type": "RobocasaGR1N15Bridge"
+            },
+            {
+              "type": "ParquetPrompter"
+            },
+            {
+              "max_len": 900,
+              "num_images": 1,
+              "tokenizer": {
+                "model_path": "fluxvla/models/third_party_models/eagle2_hg_model",
+                "type": "PretrainedTokenizer"
+              },
+              "type": "ProcessPromptsWithImage"
+            },
+            {
+              "scale": 0.95,
+              "type": "RandomCropImages"
+            },
+            {
+              "height": 224,
+              "type": "ResizeImages",
+              "width": 224
+            },
+            {
+              "brightness": 0.3,
+              "contrast": 0.4,
+              "hue": 0.08,
+              "saturation": 0.5,
+              "type": "ColorJitterImages"
+            },
+            {
+              "means": [
+                [
+                  127.5,
+                  127.5,
+                  127.5
+                ]
+              ],
+              "stds": [
+                [
+                  127.5,
+                  127.5,
+                  127.5
+                ]
+              ],
+              "type": "NormalizeImages"
+            },
+            {
+              "action_dim": 32,
+              "action_key": "action",
+              "norm_type": "min_max",
+              "normalize_states": false,
+              "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": "robocasa_gr1_24tasks_30ep",
+      "type": "DistributedRepeatingDataset"
+    },
+    "per_device_batch_size": 16,
+    "per_device_num_workers": 4
+  }
+}

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/config.yaml

@@ -0,0 +1,281 @@
+_OFFICIAL_GR1_STATS_PATH: work_dirs/official_groot_gr1_dataset_statistics.json
+_STAT: robocasa_gr1_24tasks_30ep
+eval:
+  dataset:
+    transforms:
+    - center_crop_scale: 0.95
+      embodiment_id: 24
+      img_key: video.ego_view_bg_crop_pad_res256_freq20
+      normalize: false
+      resize_size: 224
+      type: ProcessRobocasaEvalInputs
+    - type: RobocasaGR1N15Bridge
+    - action_key: action
+      norm_type: min_max
+      normalize_states: false
+      state_dim: 64
+      state_key: proprio
+      type: NormalizeStatesAndActions
+    - max_len: 900
+      num_images: 1
+      return_text: true
+      tokenizer:
+        model_path: fluxvla/models/third_party_models/eagle2_hg_model
+        type: PretrainedTokenizer
+      type: ProcessPromptsWithImage
+    - means:
+      - - 127.5
+        - 127.5
+        - 127.5
+      stds:
+      - - 127.5
+        - 127.5
+        - 127.5
+      type: NormalizeImages
+    type: RobocasaEvalDataset
+    unnorm_key: robocasa_gr1_24tasks_30ep
+  denormalize_action:
+    action_dim: 29
+    clip_actions: false
+    norm_type: min_max
+    stats_order: fluxvla
+    type: DenormalizeRobocasaAction
+  eval_chunk_size: 16
+  max_episode_steps: 720
+  model_family: groot
+  num_trials_per_task: 20
+  seed: 7
+  task_list:
+  - gr1_unified/PnPBottleToCabinetClose_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PnPCanToDrawerClose_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PnPCupToDrawerClose_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PnPMilkToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PnPPotatoToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PnPWineToCabinetClose_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromCuttingboardToBasketSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromCuttingboardToPanSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromCuttingboardToPotSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlacematToBasketSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlacematToBowlSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlacematToPlateSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlacematToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlateToBowlSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlateToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlateToPanSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromPlateToPlateSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromTrayToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromTrayToPlateSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromTrayToPotSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromTrayToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env
+  - gr1_unified/PosttrainPnPNovelFromTrayToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env
+  type: RobocasaEvalRunner
+  unnorm_key: robocasa_gr1_24tasks_30ep
+inference_model:
+  pretrained_name_or_path: ./checkpoints/GR00T-N1.5-3B
+  type: LlavaVLA
+  vla_head:
+    action_dim: 32
+    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
+    max_input_seq_len: 900
+    num_heads: 4
+    num_inference_timesteps: 4
+    num_layers: 1
+    num_steps: 16
+    ori_action_dim: 29
+    state_dim: 64
+    traj_length: 16
+    type: FlowMatchingInferenceHead
+  vlm_backbone:
+    type: EagleInferenceBackbone
+    vlm_config:
+      max_input_seq_len: 900
+    vlm_path: fluxvla/models/third_party_models/eagle2_hg_model
+model:
+  freeze_projector: false
+  freeze_vlm_backbone: false
+  name_mapping:
+    vla_head: action_head
+    vlm_backbone.vlm: backbone.eagle_model
+  pretrained_name_or_path: ./checkpoints/GR00T-N1.5-3B
+  type: LlavaVLA
+  vla_head:
+    action_dim: 32
+    hidden_size: 1024
+    input_embedding_dim: 1536
+    num_heads: 4
+    num_inference_timesteps: 4
+    num_layers: 1
+    num_steps: 16
+    ori_action_dim: 29
+    state_dim: 64
+    traj_length: 16
+    type: FlowMatchingHead
+  vlm_backbone:
+    dtype: bf16
+    tune_llm: false
+    tune_visual: true
+    type: EagleBackbone
+    vlm_config:
+      max_input_seq_len: 900
+    vlm_path: fluxvla/models/third_party_models/eagle2_hg_model
+runner:
+  change_key_name: false
+  collator:
+    keys:
+    - states
+    - observation.eepose
+    - timestamp
+    - images
+    - img_masks
+    - lang_tokens
+    - lang_masks
+    - actions
+    - action_masks
+    - embodiment_ids
+    meta_keys:
+    - task_description
+    - prompt
+    - info
+    - stats
+    type: DictCollator
+  enable_gradient_checkpointing: false
+  enable_mixed_precision_training: true
+  learning_rate: 6.0e-05
+  lr_scheduler_type: linear-warmup+cosine-decay
+  max_epochs: null
+  max_grad_norm: 1.0
+  max_keep_ckpts: 20
+  max_steps: 50000
+  metric:
+    active_trackers:
+    - jsonl
+    - wandb
+    grad_accumulation_steps: 1
+    run_dir: work_dirs
+    type: VLAMetric
+    window_size: 1
+  mixed_precision_dtype: bf16
+  sampler: null
+  save_epoch_interval: 1
+  save_full_model: true
+  save_iter_interval: 5000
+  sharding_strategy: no-shard
+  tokenizer:
+    model_path: fluxvla/models/third_party_models/eagle2_hg_model
+    type: PretrainedTokenizer
+  type: FSDPTrainRunner
+  warmup_ratio: 0.05
+  weight_decay: 1.0e-05
+train_dataloader:
+  dataset:
+    dataset_statistics_path: work_dirs/official_groot_gr1_dataset_statistics.json
+    datasets:
+    - action_key: action
+      action_window_size: 16
+      data_root_path:
+      - ./datasets/robocasa_fluxvla/PnPBottleToCabinetClose
+      - ./datasets/robocasa_fluxvla/PnPCanToDrawerClose
+      - ./datasets/robocasa_fluxvla/PnPCupToDrawerClose
+      - ./datasets/robocasa_fluxvla/PnPMilkToMicrowaveClose
+      - ./datasets/robocasa_fluxvla/PnPPotatoToMicrowaveClose
+      - ./datasets/robocasa_fluxvla/PnPWineToCabinetClose
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToBasketSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPanSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPotSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBasketSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBowlSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToPlateSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToTieredshelfSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToBowlSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToCardboardboxSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPanSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPlateSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToCardboardboxSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPlateSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPotSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredbasketSplitA
+      - ./datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredshelfSplitA
+      statistic_name: robocasa_gr1_24tasks_30ep
+      transforms:
+      - embodiment_id: 24
+        name_mappings:
+          actions:
+          - actions
+          observation.state:
+          - states
+        parquet_keys:
+        - observation.state
+        - timestamp
+        - actions
+        - info
+        - stats
+        - action_masks
+        type: ProcessParquetInputs
+        video_keys:
+        - observation.images.ego_view
+      - type: RobocasaGR1N15Bridge
+      - type: ParquetPrompter
+      - max_len: 900
+        num_images: 1
+        tokenizer:
+          model_path: fluxvla/models/third_party_models/eagle2_hg_model
+          type: PretrainedTokenizer
+        type: ProcessPromptsWithImage
+      - scale: 0.95
+        type: RandomCropImages
+      - height: 224
+        type: ResizeImages
+        width: 224
+      - brightness: 0.3
+        contrast: 0.4
+        hue: 0.08
+        saturation: 0.5
+        type: ColorJitterImages
+      - means:
+        - - 127.5
+          - 127.5
+          - 127.5
+        stds:
+        - - 127.5
+          - 127.5
+          - 127.5
+        type: NormalizeImages
+      - action_dim: 32
+        action_key: action
+        norm_type: min_max
+        normalize_states: false
+        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: robocasa_gr1_24tasks_30ep
+    type: DistributedRepeatingDataset
+  per_device_batch_size: 16
+  per_device_num_workers: 4

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/dataset_statistics.json

@@ -0,0 +1,396 @@
+{
+  "robocasa_gr1_24tasks_30ep": {
+    "proprio": {
+      "mean": [
+        -0.1721954196691514,
+        0.18329788744449615,
+        -0.1360361576080323,
+        -1.0807313919067383,
+        0.14450088143348713,
+        0.13195440173149092,
+        -0.08306947350502007,
+        0.21940384805202476,
+        0.21819256246089933,
+        0.20426052808761605,
+        0.22833923995494854,
+        0.06841608136892328,
+        0.9091449975967411,
+        -0.1590277999639513,
+        -0.2511773109436038,
+        0.08599518239498137,
+        -1.1420482397079474,
+        0.21259814500808735,
+        0.04237671568989762,
+        0.03427042067050932,
+        0.34993353486061096,
+        0.3412855267524723,
+        0.3271332383155817,
+        0.3575766980648041,
+        0.09723818302154554,
+        1.0641368627548227,
+        0.014510802924633035,
+        0.014398206025362044,
+        -0.00020241182937752458
+      ],
+      "std": [
+        0.4068214595317843,
+        0.18259239196777355,
+        0.2513278424739838,
+        0.5786943435668955,
+        0.33050763607025135,
+        0.28609949350357067,
+        0.3013432323932649,
+        0.42784574627876293,
+        0.4130409955978396,
+        0.3710329830646513,
+        0.4229725301265714,
+        0.18603332340717335,
+        0.7297042012214684,
+        0.5631562471389767,
+        0.24450603127479503,
+        0.3282549381256103,
+        0.7310536503791792,
+        0.48139148950576843,
+        0.35611921548843384,
+        0.5090531110763546,
+        0.49147421121597257,
+        0.46545258164405845,
+        0.44516301155090443,
+        0.4866269826889039,
+        0.1814946085214615,
+        0.5735033154487607,
+        0.1227026358246804,
+        0.03244836255908008,
+        0.007282953243702651
+      ],
+      "min": [
+        -1.7593636512756348,
+        -0.08899051696062088,
+        -1.6571800708770752,
+        -2.516343355178833,
+        -2.307133436203003,
+        -1.4132264852523804,
+        -1.4408576488494873,
+        -3.262805700302124,
+        -2.7771592140197754,
+        -3.4712820053100586,
+        -2.7828404903411865,
+        -2.197380781173706,
+        -2.636387825012207,
+        -2.2743337154388428,
+        -2.999122142791748,
+        -1.7424250841140747,
+        -2.536545753479004,
+        -2.6034910678863525,
+        -1.5032434463500977,
+        -1.4913568496704102,
+        -1.7595213651657104,
+        -1.8982670307159424,
+        -1.8622381687164307,
+        -1.8314869403839111,
+        -1.0045679807662964,
+        -0.31411829590797424,
+        -0.8131351470947266,
+        -0.4901888370513916,
+        -0.7304351925849915
+      ],
+      "max": [
+        1.3570959568023682,
+        1.5781080722808838,
+        1.7425614595413208,
+        0.04977008327841759,
+        2.521491289138794,
+        1.5104753971099854,
+        1.3451268672943115,
+        2.8765757083892822,
+        2.6975045204162598,
+        3.3291828632354736,
+        2.7148845195770264,
+        3.664421558380127,
+        3.5466625690460205,
+        1.7399436235427856,
+        0.07876992970705032,
+        2.5426626205444336,
+        0.048908114433288574,
+        3.000918388366699,
+        1.4513440132141113,
+        1.5000383853912354,
+        2.397510290145874,
+        2.4422738552093506,
+        3.0158462524414062,
+        2.457378625869751,
+        1.4605257511138916,
+        2.168534278869629,
+        1.0323854684829712,
+        0.7102982997894287,
+        0.43762317299842834
+      ],
+      "q01": [
+        -1.6011584222316746,
+        -0.00048318122862838164,
+        -0.9488833963871,
+        -2.4362342357635502,
+        -0.9403592050075524,
+        -0.6489510446786882,
+        -0.9641011685132993,
+        -0.17130047902464876,
+        -0.0025680272374302166,
+        -0.0029985445667989555,
+        -0.002490740863140675,
+        -0.14894675761461273,
+        -0.00027310136792948466,
+        -1.4233640372753156,
+        -1.0436977982521036,
+        -0.7169023394584662,
+        -2.4936094284057644,
+        -0.7377233326435084,
+        -0.8036172896623616,
+        -0.9991738587617871,
+        -0.004218762856908145,
+        -0.004575904295779761,
+        -0.005576773989014321,
+        -0.0047070790780708175,
+        -0.18967307358980165,
+        0.03200087323784828,
+        -0.3365072563290599,
+        -0.02904987111687659,
+        -0.02606033049523827
+      ],
+      "q99": [
+        0.6305126994848259,
+        0.7966297268867486,
+        0.3873166218399999,
+        -0.07399308793246723,
+        1.0455405175685892,
+        0.8574009954929348,
+        0.4838866069912913,
+        1.5014824867248553,
+        1.501653671264647,
+        1.4043268024921445,
+        1.5127789616584784,
+        0.781709930300714,
+        1.872669064998624,
+        1.167767405509949,
+        0.0006303858070168673,
+        0.8771779745817198,
+        -0.0582109544426202,
+        1.470731711387636,
+        0.8474899172782918,
+        1.2271583080291764,
+        1.4994217157363876,
+        1.5029035806655904,
+        1.6614551365375507,
+        1.5188306391239186,
+        0.6953653633594535,
+        1.8182492077350652,
+        0.48663658648729347,
+        0.12612193301320107,
+        0.022280203178524964
+      ]
+    },
+    "timestamp": {
+      "mean": [
+        0.0
+      ],
+      "std": [
+        1.0
+      ],
+      "min": [
+        0.0
+      ],
+      "max": [
+        0.0
+      ],
+      "q01": null,
+      "q99": null
+    },
+    "action": {
+      "mean": [
+        -0.17771673202514657,
+        0.18351773917675007,
+        -0.13720993697643286,
+        -1.0842658281326296,
+        0.14675945043563873,
+        0.12921456992626204,
+        -0.09504522383213035,
+        -0.22721463441848752,
+        -0.21956320106983204,
+        -0.2144497036933901,
+        -0.21446840465068817,
+        -0.48631641268730147,
+        1.6838233470916744,
+        -0.16953834891319267,
+        -0.2535267770290373,
+        0.082216002047062,
+        -1.1492931842803968,
+        0.2176141738891601,
+        0.04310995712876314,
+        0.04133718833327294,
+        -0.543570637702943,
+        -0.5304322242736823,
+        -0.5179098844528184,
+        -0.4968528747558592,
+        -1.1343770027160642,
+        2.128209590911863,
+        0.014836843125522158,
+        0.010428135283291354,
+        -0.00014937532250769447
+      ],
+      "std": [
+        0.4174487590789793,
+        0.18478454649448395,
+        0.25687918066978443,
+        0.5807053446769717,
+        0.33657354116439797,
+        0.29842707514762895,
+        0.31536528468132013,
+        1.0050891637802128,
+        1.0124161243438718,
+        1.0138051509857173,
+        1.0200439691543575,
+        2.066762685775758,
+        1.4588384628295925,
+        0.5752111673355106,
+        0.25846862792968733,
+        0.34115546941757197,
+        0.7394758462905866,
+        0.4870710670948032,
+        0.3721199631690978,
+        0.5423213839530941,
+        1.1248024702072141,
+        1.1392076015472405,
+        1.1426663398742682,
+        1.301862597465516,
+        2.199875354766848,
+        0.9186902642250119,
+        0.12568794190883642,
+        0.030695030465722098,
+        0.004574332851916548
+      ],
+      "min": [
+        -2.291214942932129,
+        -0.001746351015754044,
+        -2.4999990463256836,
+        -2.814260482788086,
+        -2.3611011505126953,
+        -1.4937989711761475,
+        -1.4969758987426758,
+        -1.9597030878067017,
+        -1.8635213375091553,
+        -1.9709523916244507,
+        -1.5,
+        -3.0,
+        0.0,
+        -2.318650960922241,
+        -2.9999561309814453,
+        -1.9257696866989136,
+        -2.8759055137634277,
+        -2.620600461959839,
+        -1.5001521110534668,
+        -1.4994292259216309,
+        -1.5,
+        -1.5,
+        -1.5,
+        -1.5,
+        -3.0,
+        4.470348358154297e-08,
+        -0.8292319774627686,
+        -0.5185094475746155,
+        -0.37811079621315
+      ],
+      "max": [
+        1.4113223552703857,
+        1.8901419639587402,
+        1.7754145860671997,
+        7.309383363462985e-05,
+        2.563594102859497,
+        1.5000243186950684,
+        1.4908421039581299,
+        1.5707963705062866,
+        1.646651268005371,
+        1.709236979484558,
+        2.0620639324188232,
+        3.0,
+        3.0,
+        1.7834906578063965,
+        0.0002448999439366162,
+        2.549729585647583,
+        7.45560391806066e-05,
+        3.0000460147857666,
+        1.4975632429122925,
+        1.4998434782028198,
+        1.5707963705062866,
+        1.7183797359466553,
+        3.979951858520508,
+        3.9879753589630127,
+        3.0,
+        3.0,
+        1.049119234085083,
+        0.6198405623435974,
+        0.45177245140075684
+      ],
+      "q01": [
+        -1.6055770933628097,
+        -7.429541994952163e-07,
+        -0.9837155520915981,
+        -2.480029511451721,
+        -0.9457575023174288,
+        -0.7192997813224794,
+        -1.008323037624359,
+        -1.4999999999999987,
+        -1.4999999999999987,
+        -1.4999999999999987,
+        -1.4999999999999987,
+        -2.9999999999999973,
+        0.0,
+        -1.4375487387180332,
+        -1.0682010412216194,
+        -0.7535711079835895,
+        -2.6383612155914276,
+        -0.7471688866615306,
+        -0.8604200631380092,
+        -1.0900398135185252,
+        -1.4999999999999987,
+        -1.4999999999999987,
+        -1.4999999999999987,
+        -1.4999999999999987,
+        -2.9999999999999973,
+        0.04336242545396091,
+        -0.3402548208832736,
+        -0.025130789913237084,
+        -0.016220059804618367
+      ],
+      "q99": [
+        0.6335921049118061,
+        0.8018405169248585,
+        0.3930345878005033,
+        -0.06366106234490865,
+        1.0508334636688252,
+        0.8918659299612054,
+        0.5138001590967192,
+        1.4999999999999987,
+        1.4999999999999987,
+        1.4999999999999987,
+        1.4999999999999987,
+        2.9999999999999973,
+        2.9999999999999973,
+        1.1753968834876984,
+        -9.999999974752431e-07,
+        0.8976036489009875,
+        -0.0579521823674439,
+        1.4876463234424584,
+        0.8774339169263857,
+        1.4641386926174185,
+        1.4999999999999987,
+        1.4999999999999987,
+        1.4999999999999987,
+        1.4999999999999987,
+        2.9999999999999973,
+        2.9999999999999973,
+        0.49598969072103627,
+        0.12137954644858844,
+        0.010726323025301106
+      ]
+    }
+  }
+}

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug_2026_05_26_16_59_09.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c76268dbe3cb4edf581d738f124dbd67c8e00f49a9e8ebc1e5e34a4491e51762
+size 14204515

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/run-metrics.jsonl

@@ -0,0 +1 @@
+{"hparams": "{'model': {'type': 'LlavaVLA', 'pretrained_name_or_path': './checkpoints/GR00T-N1.5-3B', 'vlm_backbone': {'type': 'EagleBackbone', 'dtype': 'bf16', 'vlm_path': 'fluxvla/models/third_party_models/eagle2_hg_model', 'vlm_config': {'max_input_seq_len': 900}, 'tune_llm': False, 'tune_visual': True}, 'vla_head': {'type': 'FlowMatchingHead', 'state_dim': 64, 'hidden_size': 1024, 'input_embedding_dim': 1536, 'num_layers': 1, 'num_heads': 4, 'num_inference_timesteps': 4, 'num_steps': 16, 'traj_length': 16, 'action_dim': 32, 'ori_action_dim': 29}, 'freeze_vlm_backbone': False, 'name_mapping': {'vlm_backbone.vlm': 'backbone.eagle_model', 'vla_head': 'action_head'}, 'freeze_projector': False}, '_STAT': 'robocasa_gr1_24tasks_30ep', '_OFFICIAL_GR1_STATS_PATH': 'work_dirs/official_groot_gr1_dataset_statistics.json', 'train_dataloader': {'per_device_batch_size': 16, 'per_device_num_workers': 4, 'dataset': {'type': 'DistributedRepeatingDataset', 'name_mappings': {'observation.state': ['proprio'], 'action': ['action']}, 'statistic_keys': ['observation.state', 'timestamp', 'action'], 'statistic_name': 'robocasa_gr1_24tasks_30ep', 'dataset_statistics_path': 'work_dirs/official_groot_gr1_dataset_statistics.json', 'datasets': [{'type': 'ParquetDataset', 'data_root_path': ['./datasets/robocasa_fluxvla/PnPBottleToCabinetClose', './datasets/robocasa_fluxvla/PnPCanToDrawerClose', './datasets/robocasa_fluxvla/PnPCupToDrawerClose', './datasets/robocasa_fluxvla/PnPMilkToMicrowaveClose', './datasets/robocasa_fluxvla/PnPPotatoToMicrowaveClose', './datasets/robocasa_fluxvla/PnPWineToCabinetClose', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToBasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPanSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPotSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBowlSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToTieredshelfSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToBowlSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPanSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPotSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredbasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredshelfSplitA'], 'statistic_name': 'robocasa_gr1_24tasks_30ep', 'action_key': 'action', 'use_delta': False, 'window_start_idx': 0, 'transforms': [{'type': 'ProcessParquetInputs', 'embodiment_id': 24, 'parquet_keys': ['observation.state', 'timestamp', 'actions', 'info', 'stats', 'action_masks'], 'video_keys': ['observation.images.ego_view'], 'name_mappings': {'observation.state': ['states'], 'actions': ['actions']}}, {'type': 'RobocasaGR1N15Bridge'}, {'type': 'ParquetPrompter'}, {'type': 'ProcessPromptsWithImage', 'max_len': 900, 'num_images': 1, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}}, {'type': 'RandomCropImages', 'scale': 0.95}, {'type': 'ResizeImages', 'height': 224, 'width': 224}, {'type': 'ColorJitterImages', 'brightness': 0.3, 'contrast': 0.4, 'saturation': 0.5, 'hue': 0.08}, {'type': 'NormalizeImages', 'means': [[127.5, 127.5, 127.5]], 'stds': [[127.5, 127.5, 127.5]]}, {'type': 'NormalizeStatesAndActions', 'state_dim': 64, 'action_dim': 32, 'state_key': 'proprio', 'action_key': 'action', 'norm_type': 'min_max', 'normalize_states': False}], 'action_window_size': 16}]}}, 'runner': {'type': 'FSDPTrainRunner', 'max_epochs': None, 'learning_rate': 6e-05, 'weight_decay': 1e-05, 'max_grad_norm': 1.0, 'sampler': None, 'save_iter_interval': 5000, 'save_epoch_interval': 1, 'max_keep_ckpts': 20, 'save_full_model': True, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}, 'collator': {'type': 'DictCollator', 'keys': ['states', 'observation.eepose', 'timestamp', 'images', 'img_masks', 'lang_tokens', 'lang_masks', 'actions', 'action_masks', 'embodiment_ids'], 'meta_keys': ['task_description', 'prompt', 'info', 'stats']}, 'metric': {'type': 'VLAMetric', 'active_trackers': ('jsonl', 'wandb'), 'run_dir': 'work_dirs/groot_robocasa_24x30_0525l_lr6e-5', 'grad_accumulation_steps': 1, 'window_size': 1, 'hparams': Config (path: configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py): {'model': {'type': 'LlavaVLA', 'pretrained_name_or_path': './checkpoints/GR00T-N1.5-3B', 'vlm_backbone': {'type': 'EagleBackbone', 'dtype': 'bf16', 'vlm_path': 'fluxvla/models/third_party_models/eagle2_hg_model', 'vlm_config': {'max_input_seq_len': 900}, 'tune_llm': False, 'tune_visual': True}, 'vla_head': {'type': 'FlowMatchingHead', 'state_dim': 64, 'hidden_size': 1024, 'input_embedding_dim': 1536, 'num_layers': 1, 'num_heads': 4, 'num_inference_timesteps': 4, 'num_steps': 16, 'traj_length': 16, 'action_dim': 32, 'ori_action_dim': 29}, 'freeze_vlm_backbone': False, 'name_mapping': {'vlm_backbone.vlm': 'backbone.eagle_model', 'vla_head': 'action_head'}, 'freeze_projector': False}, '_STAT': 'robocasa_gr1_24tasks_30ep', '_OFFICIAL_GR1_STATS_PATH': 'work_dirs/official_groot_gr1_dataset_statistics.json', 'train_dataloader': {'per_device_batch_size': 16, 'per_device_num_workers': 4, 'dataset': {'type': 'DistributedRepeatingDataset', 'name_mappings': {'observation.state': ['proprio'], 'action': ['action']}, 'statistic_keys': ['observation.state', 'timestamp', 'action'], 'statistic_name': 'robocasa_gr1_24tasks_30ep', 'dataset_statistics_path': 'work_dirs/official_groot_gr1_dataset_statistics.json', 'datasets': [{'type': 'ParquetDataset', 'data_root_path': ['./datasets/robocasa_fluxvla/PnPBottleToCabinetClose', './datasets/robocasa_fluxvla/PnPCanToDrawerClose', './datasets/robocasa_fluxvla/PnPCupToDrawerClose', './datasets/robocasa_fluxvla/PnPMilkToMicrowaveClose', './datasets/robocasa_fluxvla/PnPPotatoToMicrowaveClose', './datasets/robocasa_fluxvla/PnPWineToCabinetClose', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToBasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPanSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPotSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBowlSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToTieredshelfSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToBowlSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPanSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPotSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredbasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredshelfSplitA'], 'statistic_name': 'robocasa_gr1_24tasks_30ep', 'action_key': 'action', 'use_delta': False, 'window_start_idx': 0, 'transforms': [{'type': 'ProcessParquetInputs', 'embodiment_id': 24, 'parquet_keys': ['observation.state', 'timestamp', 'actions', 'info', 'stats', 'action_masks'], 'video_keys': ['observation.images.ego_view'], 'name_mappings': {'observation.state': ['states'], 'actions': ['actions']}}, {'type': 'RobocasaGR1N15Bridge'}, {'type': 'ParquetPrompter'}, {'type': 'ProcessPromptsWithImage', 'max_len': 900, 'num_images': 1, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}}, {'type': 'RandomCropImages', 'scale': 0.95}, {'type': 'ResizeImages', 'height': 224, 'width': 224}, {'type': 'ColorJitterImages', 'brightness': 0.3, 'contrast': 0.4, 'saturation': 0.5, 'hue': 0.08}, {'type': 'NormalizeImages', 'means': [[127.5, 127.5, 127.5]], 'stds': [[127.5, 127.5, 127.5]]}, {'type': 'NormalizeStatesAndActions', 'state_dim': 64, 'action_dim': 32, 'state_key': 'proprio', 'action_key': 'action', 'norm_type': 'min_max', 'normalize_states': False}], 'action_window_size': 16}]}}, 'runner': {'type': 'FSDPTrainRunner', 'max_epochs': None, 'learning_rate': 6e-05, 'weight_decay': 1e-05, 'max_grad_norm': 1.0, 'sampler': None, 'save_iter_interval': 5000, 'save_epoch_interval': 1, 'max_keep_ckpts': 20, 'save_full_model': True, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}, 'collator': {'type': 'DictCollator', 'keys': ['states', 'observation.eepose', 'timestamp', 'images', 'img_masks', 'lang_tokens', 'lang_masks', 'actions', 'action_masks', 'embodiment_ids'], 'meta_keys': ['task_description', 'prompt', 'info', 'stats']}, 'metric': {'type': 'VLAMetric', 'active_trackers': ('jsonl', 'wandb'), 'run_dir': 'work_dirs/groot_robocasa_24x30_0525l_lr6e-5', 'grad_accumulation_steps': 1, 'window_size': 1, 'hparams': Config (path: configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py): {...}, 'run_id': 'gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug_2026_05_26_16_59_09'}, 'lr_scheduler_type': 'linear-warmup+cosine-decay', 'warmup_ratio': 0.05, 'enable_gradient_checkpointing': False, 'enable_mixed_precision_training': True, 'mixed_precision_dtype': 'bf16', 'sharding_strategy': 'no-shard', 'change_key_name': False, 'max_steps': 50000, 'cfg': Config (path: configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py): {...}, 'args': Namespace(config='configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py', work_dir='work_dirs/groot_robocasa_24x30_0525l_lr6e-5', cfg_options={'runner.type': 'FSDPTrainRunner', 'runner.sharding_strategy': 'no-shard', 'train_dataloader.per_device_batch_size': 16, 'runner.enable_gradient_checkpointing': False, 'runner.learning_rate': 6e-05, 'runner.weight_decay': 1e-05, 'runner.warmup_ratio': 0.05, 'runner.lr_scheduler_type': 'linear-warmup+cosine-decay', 'runner.save_iter_interval': 5000, 'runner.save_epoch_interval': 1, 'runner.max_keep_ckpts': 20, 'runner.metric.active_trackers': ('jsonl', 'wandb'), 'runner.max_epochs': None, 'runner.max_steps': 50000}, eval_after_train=False, resume_from=None)}, 'inference_model': {'type': 'LlavaVLA', 'pretrained_name_or_path': './checkpoints/GR00T-N1.5-3B', 'vlm_backbone': {'type': 'EagleInferenceBackbone', 'vlm_path': 'fluxvla/models/third_party_models/eagle2_hg_model', 'vlm_config': {'max_input_seq_len': 900}}, 'vla_head': {'type': 'FlowMatchingInferenceHead', 'state_dim': 64, 'hidden_size': 1024, 'input_embedding_dim': 1536, 'num_layers': 1, 'num_heads': 4, 'num_steps': 16, 'num_inference_timesteps': 4, 'traj_length': 16, 'ori_action_dim': 29, 'action_dim': 32, 'max_input_seq_len': 900, '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': None}}}, 'eval': {'type': 'RobocasaEvalRunner', 'model_family': 'groot', 'task_list': ['gr1_unified/PnPBottleToCabinetClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPCanToDrawerClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPCupToDrawerClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPMilkToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPPotatoToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPWineToCabinetClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToBasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToPanSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToPotSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToBasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToBowlSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToBowlSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToPanSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToPotSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env'], 'eval_chunk_size': 16, 'max_episode_steps': 720, 'num_trials_per_task': 20, 'seed': 7, 'unnorm_key': 'robocasa_gr1_24tasks_30ep', 'dataset': {'type': 'RobocasaEvalDataset', 'unnorm_key': 'robocasa_gr1_24tasks_30ep', 'transforms': [{'type': 'ProcessRobocasaEvalInputs', 'img_key': 'video.ego_view_bg_crop_pad_res256_freq20', 'resize_size': 224, 'center_crop_scale': 0.95, 'normalize': False, 'embodiment_id': 24}, {'type': 'RobocasaGR1N15Bridge'}, {'type': 'NormalizeStatesAndActions', 'state_dim': 64, 'state_key': 'proprio', 'action_key': 'action', 'norm_type': 'min_max', 'normalize_states': False}, {'type': 'ProcessPromptsWithImage', 'max_len': 900, 'num_images': 1, 'return_text': True, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}}, {'type': 'NormalizeImages', 'means': [[127.5, 127.5, 127.5]], 'stds': [[127.5, 127.5, 127.5]]}]}, 'denormalize_action': {'type': 'DenormalizeRobocasaAction', 'norm_type': 'min_max', 'action_dim': 29, 'clip_actions': False, 'stats_order': 'fluxvla'}}}, 'run_id': 'gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug_2026_05_26_16_59_09'}, 'lr_scheduler_type': 'linear-warmup+cosine-decay', 'warmup_ratio': 0.05, 'enable_gradient_checkpointing': False, 'enable_mixed_precision_training': True, 'mixed_precision_dtype': 'bf16', 'sharding_strategy': 'no-shard', 'change_key_name': False, 'max_steps': 50000, 'cfg': Config (path: configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py): {'model': {'type': 'LlavaVLA', 'pretrained_name_or_path': './checkpoints/GR00T-N1.5-3B', 'vlm_backbone': {'type': 'EagleBackbone', 'dtype': 'bf16', 'vlm_path': 'fluxvla/models/third_party_models/eagle2_hg_model', 'vlm_config': {'max_input_seq_len': 900}, 'tune_llm': False, 'tune_visual': True}, 'vla_head': {'type': 'FlowMatchingHead', 'state_dim': 64, 'hidden_size': 1024, 'input_embedding_dim': 1536, 'num_layers': 1, 'num_heads': 4, 'num_inference_timesteps': 4, 'num_steps': 16, 'traj_length': 16, 'action_dim': 32, 'ori_action_dim': 29}, 'freeze_vlm_backbone': False, 'name_mapping': {'vlm_backbone.vlm': 'backbone.eagle_model', 'vla_head': 'action_head'}, 'freeze_projector': False}, '_STAT': 'robocasa_gr1_24tasks_30ep', '_OFFICIAL_GR1_STATS_PATH': 'work_dirs/official_groot_gr1_dataset_statistics.json', 'train_dataloader': {'per_device_batch_size': 16, 'per_device_num_workers': 4, 'dataset': {'type': 'DistributedRepeatingDataset', 'name_mappings': {'observation.state': ['proprio'], 'action': ['action']}, 'statistic_keys': ['observation.state', 'timestamp', 'action'], 'statistic_name': 'robocasa_gr1_24tasks_30ep', 'dataset_statistics_path': 'work_dirs/official_groot_gr1_dataset_statistics.json', 'datasets': [{'type': 'ParquetDataset', 'data_root_path': ['./datasets/robocasa_fluxvla/PnPBottleToCabinetClose', './datasets/robocasa_fluxvla/PnPCanToDrawerClose', './datasets/robocasa_fluxvla/PnPCupToDrawerClose', './datasets/robocasa_fluxvla/PnPMilkToMicrowaveClose', './datasets/robocasa_fluxvla/PnPPotatoToMicrowaveClose', './datasets/robocasa_fluxvla/PnPWineToCabinetClose', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToBasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPanSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToPotSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToBowlSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlacematToTieredshelfSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToBowlSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPanSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromPlateToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToCardboardboxSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPlateSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToPotSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredbasketSplitA', './datasets/robocasa_fluxvla/PosttrainPnPNovelFromTrayToTieredshelfSplitA'], 'statistic_name': 'robocasa_gr1_24tasks_30ep', 'action_key': 'action', 'use_delta': False, 'window_start_idx': 0, 'transforms': [{'type': 'ProcessParquetInputs', 'embodiment_id': 24, 'parquet_keys': ['observation.state', 'timestamp', 'actions', 'info', 'stats', 'action_masks'], 'video_keys': ['observation.images.ego_view'], 'name_mappings': {'observation.state': ['states'], 'actions': ['actions']}}, {'type': 'RobocasaGR1N15Bridge'}, {'type': 'ParquetPrompter'}, {'type': 'ProcessPromptsWithImage', 'max_len': 900, 'num_images': 1, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}}, {'type': 'RandomCropImages', 'scale': 0.95}, {'type': 'ResizeImages', 'height': 224, 'width': 224}, {'type': 'ColorJitterImages', 'brightness': 0.3, 'contrast': 0.4, 'saturation': 0.5, 'hue': 0.08}, {'type': 'NormalizeImages', 'means': [[127.5, 127.5, 127.5]], 'stds': [[127.5, 127.5, 127.5]]}, {'type': 'NormalizeStatesAndActions', 'state_dim': 64, 'action_dim': 32, 'state_key': 'proprio', 'action_key': 'action', 'norm_type': 'min_max', 'normalize_states': False}], 'action_window_size': 16}]}}, 'runner': {'type': 'FSDPTrainRunner', 'max_epochs': None, 'learning_rate': 6e-05, 'weight_decay': 1e-05, 'max_grad_norm': 1.0, 'sampler': None, 'save_iter_interval': 5000, 'save_epoch_interval': 1, 'max_keep_ckpts': 20, 'save_full_model': True, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}, 'collator': {'type': 'DictCollator', 'keys': ['states', 'observation.eepose', 'timestamp', 'images', 'img_masks', 'lang_tokens', 'lang_masks', 'actions', 'action_masks', 'embodiment_ids'], 'meta_keys': ['task_description', 'prompt', 'info', 'stats']}, 'metric': {'type': 'VLAMetric', 'active_trackers': ('jsonl', 'wandb'), 'run_dir': 'work_dirs/groot_robocasa_24x30_0525l_lr6e-5', 'grad_accumulation_steps': 1, 'window_size': 1, 'hparams': Config (path: configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py): {...}, 'run_id': 'gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug_2026_05_26_16_59_09'}, 'lr_scheduler_type': 'linear-warmup+cosine-decay', 'warmup_ratio': 0.05, 'enable_gradient_checkpointing': False, 'enable_mixed_precision_training': True, 'mixed_precision_dtype': 'bf16', 'sharding_strategy': 'no-shard', 'change_key_name': False, 'max_steps': 50000, 'cfg': Config (path: configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py): {...}, 'args': Namespace(config='configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py', work_dir='work_dirs/groot_robocasa_24x30_0525l_lr6e-5', cfg_options={'runner.type': 'FSDPTrainRunner', 'runner.sharding_strategy': 'no-shard', 'train_dataloader.per_device_batch_size': 16, 'runner.enable_gradient_checkpointing': False, 'runner.learning_rate': 6e-05, 'runner.weight_decay': 1e-05, 'runner.warmup_ratio': 0.05, 'runner.lr_scheduler_type': 'linear-warmup+cosine-decay', 'runner.save_iter_interval': 5000, 'runner.save_epoch_interval': 1, 'runner.max_keep_ckpts': 20, 'runner.metric.active_trackers': ('jsonl', 'wandb'), 'runner.max_epochs': None, 'runner.max_steps': 50000}, eval_after_train=False, resume_from=None)}, 'inference_model': {'type': 'LlavaVLA', 'pretrained_name_or_path': './checkpoints/GR00T-N1.5-3B', 'vlm_backbone': {'type': 'EagleInferenceBackbone', 'vlm_path': 'fluxvla/models/third_party_models/eagle2_hg_model', 'vlm_config': {'max_input_seq_len': 900}}, 'vla_head': {'type': 'FlowMatchingInferenceHead', 'state_dim': 64, 'hidden_size': 1024, 'input_embedding_dim': 1536, 'num_layers': 1, 'num_heads': 4, 'num_steps': 16, 'num_inference_timesteps': 4, 'traj_length': 16, 'ori_action_dim': 29, 'action_dim': 32, 'max_input_seq_len': 900, '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': None}}}, 'eval': {'type': 'RobocasaEvalRunner', 'model_family': 'groot', 'task_list': ['gr1_unified/PnPBottleToCabinetClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPCanToDrawerClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPCupToDrawerClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPMilkToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPPotatoToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPWineToCabinetClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToBasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToPanSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToPotSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToBasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToBowlSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToBowlSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToPanSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToPotSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env'], 'eval_chunk_size': 16, 'max_episode_steps': 720, 'num_trials_per_task': 20, 'seed': 7, 'unnorm_key': 'robocasa_gr1_24tasks_30ep', 'dataset': {'type': 'RobocasaEvalDataset', 'unnorm_key': 'robocasa_gr1_24tasks_30ep', 'transforms': [{'type': 'ProcessRobocasaEvalInputs', 'img_key': 'video.ego_view_bg_crop_pad_res256_freq20', 'resize_size': 224, 'center_crop_scale': 0.95, 'normalize': False, 'embodiment_id': 24}, {'type': 'RobocasaGR1N15Bridge'}, {'type': 'NormalizeStatesAndActions', 'state_dim': 64, 'state_key': 'proprio', 'action_key': 'action', 'norm_type': 'min_max', 'normalize_states': False}, {'type': 'ProcessPromptsWithImage', 'max_len': 900, 'num_images': 1, 'return_text': True, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}}, {'type': 'NormalizeImages', 'means': [[127.5, 127.5, 127.5]], 'stds': [[127.5, 127.5, 127.5]]}]}, 'denormalize_action': {'type': 'DenormalizeRobocasaAction', 'norm_type': 'min_max', 'action_dim': 29, 'clip_actions': False, 'stats_order': 'fluxvla'}}}, 'args': Namespace(config='configs/gr00t/gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug.py', work_dir='work_dirs/groot_robocasa_24x30_0525l_lr6e-5', cfg_options={'runner.type': 'FSDPTrainRunner', 'runner.sharding_strategy': 'no-shard', 'train_dataloader.per_device_batch_size': 16, 'runner.enable_gradient_checkpointing': False, 'runner.learning_rate': 6e-05, 'runner.weight_decay': 1e-05, 'runner.warmup_ratio': 0.05, 'runner.lr_scheduler_type': 'linear-warmup+cosine-decay', 'runner.save_iter_interval': 5000, 'runner.save_epoch_interval': 1, 'runner.max_keep_ckpts': 20, 'runner.metric.active_trackers': ('jsonl', 'wandb'), 'runner.max_epochs': None, 'runner.max_steps': 50000}, eval_after_train=False, resume_from=None)}, 'inference_model': {'type': 'LlavaVLA', 'pretrained_name_or_path': './checkpoints/GR00T-N1.5-3B', 'vlm_backbone': {'type': 'EagleInferenceBackbone', 'vlm_path': 'fluxvla/models/third_party_models/eagle2_hg_model', 'vlm_config': {'max_input_seq_len': 900}}, 'vla_head': {'type': 'FlowMatchingInferenceHead', 'state_dim': 64, 'hidden_size': 1024, 'input_embedding_dim': 1536, 'num_layers': 1, 'num_heads': 4, 'num_steps': 16, 'num_inference_timesteps': 4, 'traj_length': 16, 'ori_action_dim': 29, 'action_dim': 32, 'max_input_seq_len': 900, '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': None}}}, 'eval': {'type': 'RobocasaEvalRunner', 'model_family': 'groot', 'task_list': ['gr1_unified/PnPBottleToCabinetClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPCanToDrawerClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPCupToDrawerClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPMilkToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPPotatoToMicrowaveClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PnPWineToCabinetClose_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToBasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToPanSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToPotSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromCuttingboardToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToBasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToBowlSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlacematToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToBowlSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToPanSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromPlateToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToCardboardboxSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToPlateSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToPotSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToTieredbasketSplitA_GR1ArmsAndWaistFourierHands_Env', 'gr1_unified/PosttrainPnPNovelFromTrayToTieredshelfSplitA_GR1ArmsAndWaistFourierHands_Env'], 'eval_chunk_size': 16, 'max_episode_steps': 720, 'num_trials_per_task': 20, 'seed': 7, 'unnorm_key': 'robocasa_gr1_24tasks_30ep', 'dataset': {'type': 'RobocasaEvalDataset', 'unnorm_key': 'robocasa_gr1_24tasks_30ep', 'transforms': [{'type': 'ProcessRobocasaEvalInputs', 'img_key': 'video.ego_view_bg_crop_pad_res256_freq20', 'resize_size': 224, 'center_crop_scale': 0.95, 'normalize': False, 'embodiment_id': 24}, {'type': 'RobocasaGR1N15Bridge'}, {'type': 'NormalizeStatesAndActions', 'state_dim': 64, 'state_key': 'proprio', 'action_key': 'action', 'norm_type': 'min_max', 'normalize_states': False}, {'type': 'ProcessPromptsWithImage', 'max_len': 900, 'num_images': 1, 'return_text': True, 'tokenizer': {'type': 'PretrainedTokenizer', 'model_path': 'fluxvla/models/third_party_models/eagle2_hg_model'}}, {'type': 'NormalizeImages', 'means': [[127.5, 127.5, 127.5]], 'stds': [[127.5, 127.5, 127.5]]}]}, 'denormalize_action': {'type': 'DenormalizeRobocasaAction', 'norm_type': 'min_max', 'action_dim': 29, 'clip_actions': False, 'stats_order': 'fluxvla'}}}", "run_id": "gr00t_eagle_3b_robocasa_posttrain_24x30_official_aug_2026_05_26_16_59_09"}

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/tokenizer/added_tokens.json

@@ -0,0 +1,39 @@
+{
+  "</box>": 151673,
+  "</img>": 151671,
+  "</interval>": 151679,
+  "</quad>": 151675,
+  "</ref>": 151677,
+  "</think>": 151668,
+  "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<IMG_CONTEXT>": 151669,
+  "<box>": 151672,
+  "<img>": 151670,
+  "<interval>": 151678,
+  "<quad>": 151674,
+  "<ref>": 151676,
+  "<think>": 151667,
+  "<tool_call>": 151657,
+  "<tool_response>": 151665,
+  "<|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_eagle_3b_robocasa_gr1_24x30_finetune_bs64/tokenizer/special_tokens_map.json

@@ -0,0 +1,42 @@
+{
+  "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|>",
+    "<IMG_CONTEXT>",
+    "<img>",
+    "</img>",
+    "<box>",
+    "</box>",
+    "<quad>",
+    "</quad>",
+    "<ref>",
+    "</ref>",
+    "<interval>",
+    "</interval>"
+  ],
+  "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_eagle_3b_robocasa_gr1_24x30_finetune_bs64/tokenizer/tokenizer_config.json

@@ -0,0 +1,344 @@
+{
+  "add_bos_token": false,
+  "add_eos_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": "<IMG_CONTEXT>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<img>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "</img>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<box>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "</box>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<quad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "</quad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151676": {
+      "content": "<ref>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151677": {
+      "content": "</ref>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151678": {
+      "content": "<interval>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151679": {
+      "content": "</interval>",
+      "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|>",
+    "<IMG_CONTEXT>",
+    "<img>",
+    "</img>",
+    "<box>",
+    "</box>",
+    "<quad>",
+    "</quad>",
+    "<ref>",
+    "</ref>",
+    "<interval>",
+    "</interval>"
+  ],
+  "auto_map": {
+    "AutoProcessor": "processing_eagle2_5_vl.Eagle2_5_VLProcessor"
+  },
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {{- messages[0].content + '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n    {%- set index = (messages|length - 1) - loop.index0 %}\n    {%- if ns.multi_step_tool and message.role == \"user\" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n        {%- set ns.multi_step_tool = false %}\n        {%- set ns.last_query_index = index %}\n    {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {%- set content = message.content %}\n        {%- set reasoning_content = '' %}\n        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}\n            {%- set reasoning_content = message.reasoning_content %}\n        {%- else %}\n            {%- if '</think>' in message.content %}\n                {%- set content = message.content.split('</think>')[-1].lstrip('\\n') %}\n                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n            {%- endif %}\n        {%- endif %}\n        {%- if loop.index0 > ns.last_query_index %}\n            {%- if loop.last or (not loop.last and reasoning_content) %}\n                {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n            {%- else %}\n                {{- '<|im_start|>' + message.role + '\\n' + content }}\n            {%- endif %}\n        {%- else %}\n            {{- '<|im_start|>' + message.role + '\\n' + content }}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n    {%- if enable_thinking is defined and enable_thinking is false %}\n        {{- '<think>\\n\\n</think>\\n\\n' }}\n    {%- endif %}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 16384,
+  "pad_token": "<|endoftext|>",
+  "processor_class": "Eagle2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

gr00t_eagle_3b_robocasa_gr1_24x30_finetune_bs64/vlm_backbone_config.json

@@ -0,0 +1,106 @@
+{
+  "_attn_implementation": "flash_attention_2",
+  "architectures": [
+    "Eagle2_5_VLForConditionalGeneration"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_eagle2_5_vl.Eagle2_5_VLConfig",
+    "AutoModel": "modeling_eagle2_5_vl.Eagle2_5_VLForConditionalGeneration"
+  },
+  "downsample_ratio": 0.5,
+  "dynamic_image_size": true,
+  "force_image_size": 224,
+  "image_token_index": 151669,
+  "initializer_range": 0.02,
+  "loss_version": "efficient_v2_cp_head",
+  "max_dynamic_tiles": 12,
+  "min_dynamic_tiles": 1,
+  "mlp_checkpoint": false,
+  "mlp_connector_layers": 1,
+  "model_type": "eagle_2_5_vl",
+  "output_attentions": false,
+  "pad2square": false,
+  "select_layer": -1,
+  "template": "qwen3-chat",
+  "text_config": {
+    "_name_or_path": "Qwen/Qwen3-1.7B",
+    "architectures": [
+      "Qwen3ForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0,
+    "bos_token_id": 151643,
+    "eos_token_id": 151645,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 2048,
+    "initializer_range": 0.02,
+    "intermediate_size": 6144,
+    "layer_types": [
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention",
+      "full_attention"
+    ],
+    "max_position_embeddings": 40960,
+    "max_window_layers": 28,
+    "model_type": "qwen3",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 12,
+    "num_key_value_heads": 8,
+    "rms_norm_eps": 1e-06,
+    "rope_scaling": null,
+    "rope_theta": 1000000,
+    "sliding_window": null,
+    "tie_word_embeddings": true,
+    "torch_dtype": "bfloat16",
+    "use_cache": false,
+    "use_sliding_window": false,
+    "vocab_size": 151680
+  },
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": null,
+  "use_backbone_lora": 0,
+  "use_llm_lora": 0,
+  "use_pixel_shuffle": false,
+  "use_thumbnail": true,
+  "vision_config": {
+    "attention_dropout": 0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "image_size": 224,
+    "intermediate_size": 4304,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 27,
+    "patch_size": 14,
+    "torch_dtype": "bfloat16"
+  }
+}