feat(policies): add Gr00t N1.7 policy (#3922)

* Add GR00T N1.7 support Add GR00T N1.7 policy configuration, checkpoint compatibility, processor parity, LIBERO documentation, and focused tests. Co-authored-by: Ryan Halabi <ryhalabi@nvidia.com> * Move Groot processor compatibility into Groot loader * Restore GR00T Flash Attention install guidance * Allow Groot fake RTC chunk prefetch * Fix GR00T N1.7 RTC action decoding * Trim GR00T N1.7 RTC chunks to valid horizon * Ignore padded GR00T N1.7 RTC prefix rows * removed n1.5 dependency * removed remaining N1.5 traces * groot: auto-enable LIBERO gripper action transform for libero_sim GR00T N1.7 emits gripper in [0,1] but LIBERO expects [-1,1]. The decode transform existed but was never auto-enabled for embodiment_tag=libero_sim, so the policy scored 0% on LIBERO eval. Auto-set it in __post_init__ (still overridable). LIBERO Spatial eval: 0% -> 98%. * Reconnect GR00T relative action processors * groot: remove dead N1.5 code (eagle2_hg_model, flow_matching_action_head, action_encoder) N1.7 backbone is nvidia/Cosmos-Reason2-2B via Qwen3VLForConditionalGeneration, not Eagle2 — eagle2_hg_model/ had zero refs outside its own dir. GR00TN17ActionHead (groot_n1_7.py) re-implements MultiEmbodimentActionEncoder + CategorySpecificLinear + swish + SinusoidalPositionalEncoding locally, so flow_matching_action_head.py (N1.5 FlowmatchingActionHead) and its sole dependency action_encoder.py are dead. Verified: no src/ or tests/ reference. Removed (~2037 LOC): - eagle2_hg_model/ (4 files, ~1575 LOC) - action_head/flow_matching_action_head.py (408 LOC) - action_head/action_encoder.py (54 LOC) cross_attention_dit.py KEPT (DiT/AlternateVLDiT/SelfAttentionTransformer live in N1.7). * groot: reuse lerobot get_device_from_parameters instead of inline lookup modeling_groot.py duplicated next(self.parameters()).device twice. LeRobot ships get_device_from_parameters in policies/utils.py (used by diffusion, vqbet, tdmpc, gaussian_actor). Reuse it for consistency with the framework. * groot: fix stale Eagle VLM docstring in processor (N1.7 uses Qwen3-VL backbone) Addresses checker nit: processor_groot.py docstring still described the N1.5 Eagle VLM path with eagle_content/eagle_* keys that no longer exist in the code. * test(groot): add N1.7 original-vs-LeRobot output parity test Verifies the LeRobot GR00T N1.7 integration produces equivalent raw action_pred to NVIDIA Isaac-GR00T for the same checkpoint, inputs, seed, precision (fp32) and attention kernel (SDPA): max|diff|=8.9e-7 on the libero_sim embodiment (GR00T-N1.7-LIBERO/libero_10). The two impls pin incompatible transformers majors (orig 4.57.3 vs LeRobot 5.x) and cannot share a process, so the original outputs + exact collated inputs are produced out-of-process and loaded from an .npz. The test skips on CI / when the checkpoint or artifact are absent. * test(groot): parametrize N1.7 parity across all checkpoint embodiments Generalize the original-vs-LeRobot N1.7 output-parity test from a single libero_sim case to every embodiment tag in the checkpoint (libero_sim, oxe_droid, real_g1, the real_r1_pro_sharpa family, and the xdof family). Inputs are built generically from checkpoint metadata; the test discovers per-tag .npz artifacts and runs one parametrized case each, loading the LeRobot model once via a fixture. All 9 embodiments match the original to fp32 epsilon (max|diff| < 3e-6), confirming the integration is correct across the model's full embodiment space and not overfit to libero_sim. * test(groot): self-contained parity test + in-repo producer + docs - Rename test_groot_n1_7_vs_original.py -> test_groot_vs_original.py - Make the test self-contained: producer script (dump_original_n1_7.py) now lives next to the test; default artifact dir is repo-relative (tests/policies/groot/artifacts/), overridable via GROOT_N1_7_PARITY_DIR. The test only reads artifacts and skips if absent -- it never creates external dirs. - Heavy .npz artifacts (~6-9MB each) are gitignored and regenerated by the producer; never committed. - Drop the verbose 'MULTIPLE EMBODIMENTS' docstring block (kept a one-line note). - Document the parity procedure in the groot policy README (docs/source/policy_groot_README.md). - Rename test fn test_groot_n1_7_get_action_parity -> test_groot_get_action_parity. 9/9 embodiments still pass (max|diff| < 3e-6, fp32 eps). * docs(groot): drop WHY TWO ENVIRONMENTS block from parity test docstring * test(groot): move parity producer into utils/ package Mirror the tests/policies/pi0_pi05/utils convention: move dump_original_n1_7.py into a tests/policies/groot/utils/ package (with __init__.py) and update all path references in the test docstring/skip-message and the policy README. * test(groot): adopt test_groot_lerobot for GR00T N1.7, drop N1.5 The test loaded MODEL_PATH='aractingi/bimanual-handover-groot-10k', an N1.5 checkpoint (config base_model_path=nvidia/GR00T-N1.5-3B, no model_version). On load, model_version defaults to n1.7 while the base path infers n1.5, so the version-consistency guard in GrootConfig.__post_init__ raised ValueError and both test_lerobot_groot_inference and test_lerobot_groot_forward_pass failed. N1.5 is no longer a supported model_version. Adopt the test for N1.7: - MODEL_PATH -> nvidia/GR00T-N1.7-3B (root-level sharded safetensors; loads via GrootPolicy.from_pretrained as a base N1.7 model). - Embodiment tag 'gr1' (N1.5) -> 'gr1_unified' (valid N1.7 tag from the checkpoint embodiment_id.json), via a single EMBODIMENT_TAG constant. - DUMMY_ACTION_HORIZON 16 -> 40 to match N1.7's native action-chunk size. - Docstrings/labels updated to 'GR00T N1.7'. Both tests run and pass on CUDA; full tests/policies/groot/ suite is 73 passed / 0 failed / 0 skipped. * docs(groot): document the N1.5 removal and the N1.7 parity test - groot.mdx: breaking-change warning and migration path (pin lerobot==0.5.1 to keep N1.5, or move to N1.7); the dead `huggingface-cli download` is replaced with `hf download`. - policy_groot_README.md: N1.5 removal note, updated paper / model-card links, and the two-comparison (model parity + preprocessor parity) description of the original-vs-LeRobot test, including the raw-observation artifacts and recorded seed. * fix(groot): N1.7 backbone loading and DiT parameter-count logging - select_layer default tracks the N1.7-3B checkpoint value (16); real checkpoint loads still override it from config.json. - get_backbone_cls recognizes Cosmos-Reason2 / Qwen3-VL backbones by name and warns (instead of silently assuming) when an unrecognized backbone is loaded only on the strength of backbone_model_type='qwen'. - 'revision' pins the GR00T checkpoint repo only and is no longer forwarded into the unrelated backbone repo load; pin the backbone via transformers_loading_kwargs instead. - DiT / SelfAttentionTransformer parameter counts go through logging.debug instead of print(). * fix(groot): N1.7 config defaults, N1.5 rejection, and processor/model runtime fixes Covers the GR00T N1.7 source trio (configuration, processor, model wrapper). Config: - GrootConfig defaults are the N1.7 values; explicitly passed legacy N1.5-era values (chunk_size=50, max_state_dim=64, ...) are remapped with a warning instead of silently. - action_decode_transform gains an 'auto' sentinel so an explicit 'none' opt-out wins over the libero_sim default and survives save/load round-trips. - action_delta_indices is cached on the inputs that determine it. - Legacy N1.5 checkpoints/configs (tokenizer_assets_repo, model_type/ architectures/eagle backbone markers) are rejected with a single clear error pointing to lerobot==0.5.1. Processor: - GrootN17ActionDecodeStep handles the 2-D (B, D) actions delivered by sync select_action (relative eef/non-eef decode in eval/record flows). - Postprocessor falls back to dataset stats when a raw checkpoint lacks the configured embodiment tag; raw-state cache is per-instance, not process-global; caller overrides (device, rename_map) are honored on the raw-checkpoint branch. - Camera/modality-key mismatches warn (including the zero-match fallback); deprecated Qwen2VLImageProcessorFast replaced with Qwen2VLImageProcessor; removed N1.5 processor steps are stubbed to raise the removal guidance and the action-unpack step is re-registered as _v2. Model: - Flash-attention probe is diagnostic-only; forward raises on a missing loss; print() replaced with logging; N1.5 base-path mismatch includes the removal guidance. * fix(groot): skip normalization overrides for training * fix(groot): GPU/tensor N1.7 image preprocessing + resize to trained resolution GR00T training was dataloader-bound (0->100->0 GPU-utilization sawtooth). GrootN17VLMEncodeStep ran the Qwen3-VL image processor per frame on PIL images on the single CPU main-loop thread, and that cost is timed inside dataloading_s (preprocessor(batch) runs in the main process, not the dataloader workers), so adding workers cannot hide it. - Feed the torchvision-backed Qwen3-VL processor (C,H,W) uint8 tensors instead of a per-frame Image.fromarray PIL roundtrip, and run resize/normalize/patchify on config.device (GPU) when available. Bit-identical on CPU when no resize is configured; with a resize only the PIL->torchvision bicubic backend differs (<2/255 per pixel). The use_albumentations path stays PIL/cv2; reload on a box without the saved device falls back to CPU. - Default image_target_size/crop to the N1.7 backbone's training geometry (256x256 / 230x230) when a checkpoint ships no image sizing (checkpoint_assets is None, e.g. finetuning nvidia/GR00T-N1.7-3B via repo-id with a new embodiment). Previously image_target_size=None disabled the resize, so full-resolution frames were patchified into ~4.7x more vision tokens than the model was trained on -- inflating dataloading_s (patchify) and update_s (VLM sequence) and skewing the input distribution. Checkpoints that pin their own sizing are honored; the default constants are shared with GR00T_N1_7_DEFAULTS. Net: preprocessing leaves the CPU critical path and the VLM sees the resolution it was trained on -- faster training/inference and a correct train/serve distribution. Affects inference too (shared preprocessor); existing checkpoints still load (backward compatible) but must be retrained to gain the benefits. * refactor(groot): N1.7 style cleanup (utils, imports, flash-attn, config) Mechanical refactor of the GR00T N1.7 policy to match the repo's architecture and style standards. No change to policy algorithm/numerics; only UX/CLI and packaging changes. Tests are intentionally left untouched (out of scope) and need updating for the removed `model_version` field. Cleanup & consolidation: - Add `groot/utils.py` holding the pure, side-effect-free helpers (JSON I/O, value coercion, stat flattening, rot6d/SE3 math, language/batch prep) shared by the config and processor layers. - Remove dead code: the unused `resolve_groot_n1_7_backbone_model` cache-resolver cluster, `GR00TN17Config.to_filtered_dict/json`, and the `_copy_default` wrapper. Imports & execution guards: - Hoist nested imports to module top; relative imports within the package, absolute for external modules. The version-gated Qwen3-VL classes import under the single `_transformers_available` guard (transformers is pinned >=5.4, which ships them). - No import-time side effects: `_register_with_transformers()` now runs in `GR00TN17.__init__` (idempotent via `register(exist_ok=True)`), and the N1.5 step stubs register lazily before pipeline deserialization (idempotent via the registry, no run-once globals). - Gate optional deps at the point of use with `require_package(..., extra="groot")`. Dependencies & docs: - Drop `flash-attn` (and its build-only dep `ninja`) from the `groot` extra; default to SDPA (numerically equivalent) with opt-in via `--policy.use_flash_attention`. Un-comment `lerobot[groot]` in the `all` extra and regenerate `uv.lock`. - Rewrite the `groot.mdx` install section: flash-attn is a purely optional, user-managed optimization that LeRobot neither installs nor requires. Config & CLI: - Surface previously-frozen knobs on `GrootConfig` (plumbed into `GR00TN17Config`; no-ops at their defaults): inference — `num_inference_timesteps`, `rtc_ramp_rate`, `use_flash_attention`; fine-tuning — `tune_top_llm_layers` (partial-LLM tuning) and `tune_vlln` (previously hardwired to True). - Convert the single-valued `model_version` and `n1_7_backbone_model` fields to internal constants. - Keep `base_model_path`: it is NOT equivalent to `pretrained_path` (raw NVIDIA checkpoints have no LeRobot `type` field and load only via `base_model_path`) and is genuinely user-tunable. - Keep the deprecated Isaac-GR00T/N1.5 fields (and the dead LoRA fields) as a back-compat block so a v0.5.1 N1.5 `config.json` still parses under draccus and is rejected with the friendly N1.5 removal message instead of an opaque decode error. * Optimize GR00T N1.7 image preprocessing * Remove PIL fallback from GR00T preprocessing * Fix GROOT relative action training stats * Address GROOT relative action review feedback * Fix GROOT N1.7 relative action stats * Fix GROOT relative action training stats * Fix GROOT relative action padding and RTC leftovers * Reset rollout state after robot episode end * Revert "Reset rollout state after robot episode end" This reverts commit 1322f45aec. * Move GROOT relative stats out of train script * Guard GR00T relative action stepwise decode * Match GR00T N1.7 OSS preprocessing and relative actions * Apply LIBERO action decode override after loading * Format GR00T OSS parity changes * chore(policies): add guards, warnings and comments + recover tests n1.5 check * fix(style): pre-commit * fix(ci): guard dependecy checks * chore(groot): move cv2 to the top as its in the default install tag * chore(policies): add explicit dataset dependecy to gr00t implementation * fix(test): add guard * fix(groot): make N1.7 letterbox opt-in * feat(groot): activate checkpoint-configured N1.7 raw-state dropout during training Isaac-GR00T applies dual state regularization during fine-tuning: raw-state zeroing driven by the processor sidecar's state_dropout_prob (0.2 for the inspected N1.7 checkpoint) plus encoded-feature dropout. Baseline LeRobot kept the processor in deterministic mode, so the raw-state dropout never activated (RCA Tier-2 contributor to the LeRobot-trained SO-101 failures). - GrootN17PackInputsStep: runtime-only 'training' flag + state_dropout_prob; whole-sample state zeroing gated on torch.is_grad_enabled() so eval and no_grad validation paths are unaffected - sidecar loader reads state_dropout_prob from processor_config.json - state_dropout_prob serializes with the step; the training flag intentionally does not (reloaded pipelines default to eval, re-enabled only when processors are rebuilt with dataset_meta) - _set_groot_preprocessor_training toggles any dataclass step exposing a 'training' field on serialized-pipeline reloads Verification: tests/policies/groot/test_groot_state_dropout.py (4 passed) on RTX PRO 6000 / CUDA 13.3. * fix(groot): align N1.7 fine-tuning optimizer/scheduler/precision with Isaac-GR00T Evidence from the LeRobot-vs-OSS checkpoint comparison: the LeRobot/HF 8k checkpoint's DiT moved only ~19% as far from base as the OSS-trained one (0.0547 vs 0.285 relative L2) - undertrained because the scheduler decayed over a hardcoded 10k steps regardless of --steps, on top of beta1/clip mismatches. - AdamW betas (0.95, 0.999) -> (0.9, 0.999) and grad_clip_norm 10.0 -> 1.0 (Isaac defaults) - scheduler: hardcoded CosineDecayWithWarmup(10k decay, floor 10% peak) -> DiffuserSchedulerConfig HF cosine with ceil(max_steps * warmup_ratio) warmup, deriving num_training_steps from the outer --steps at runtime - model_params_fp32 (default true): keep master weights in FP32 and compute under BF16 autocast like the native N1.7 recipe (fixes optimizer-update numerics vs pure-BF16 params) - weight-decay grouping via transformers get_parameter_names: biases and norm parameters excluded from decay - restore the TF4 lm_head/embedding weight tie so the unused Qwen LM head stays frozen and deduplicated in checkpoints - action_mask kept in native dtype for the masked flow-matching loss - drop_n_last_frames: exclude episode tails that cannot supply a complete action chunk (Isaac sampler behavior) Verification: tests/policies/groot/test_groot_training_optim_contract.py (7 passed) + remaining groot suite 11 passed/5 skipped on RTX PRO 6000 / CUDA 13.3. Note: tests/policies/groot/test_groot_n1_7.py does not collect on the base branch (pre-existing ImportError, fixed in PR #37). * feat(groot): train-time random crop for N1.7 (eval keeps center crop) Isaac-GR00T crops a random crop_fraction window during training and the deterministic center window at eval, replaying the sampled window across all camera views of a sample. This contract is unchanged since the N1.5 release (gr00t/data/transform/video.py: "If mode is 'train', return a random crop transform. If mode is 'eval', return a center crop transform.") and mirrors LeRobot's own Diffusion/VQBeT crop_is_random pattern. The LeRobot N1.7 port used the eval center crop for training too, so the fine-tuned projector/DiT never sees frame borders and trains on a single fixed appearance point. Scope: crop geometry ONLY - no color jitter, no new dependencies. The random window is plain numpy slicing inside the existing cv2 eval transform: - _transform_n1_7_image_for_vlm_albumentations gains crop_position=(y, x) fractions; None keeps the center crop byte-identical to before (verified by test) - GrootN17VLMEncodeStep gains a runtime-only 'training' flag (never serialized; reloaded pipelines default to eval); training samples ONE window per sample and reuses it across (timestep, view) frames - Isaac's cross-view consistency - gated on torch.is_grad_enabled() so no_grad validation and frozen-eval paths are unaffected - wired via dataset_meta is not None in make_groot_pre_post_processors and the existing _set_groot_preprocessor_training on serialized reloads Verification: tests/policies/groot/test_groot_train_random_crop.py (8 passed: center-crop bit-exactness with crop_position=None, corner/center windows, cross-view replay, train!=eval, no_grad gating, seed reproducibility, serialization contract) + groot suite 23 passed / 5 skipped on RTX PRO 6000 / CUDA 13.3. * docs(groot): update Training & hardware Evaluation commands Replace the multi-GPU accelerate-launch Training snippet with the current single-command 'uv run lerobot-train' N1.7 recipe (relative actions excluding gripper, bf16, flash attention, chunk/n_action_steps=16, bs64/20k steps). Replace the bimanual 'Evaluate in your hardware setup' rollout example with the SO-101 follower RTC 'uv run lerobot-rollout' command (strategy.type=base, inference.type=rtc, wrist+front cameras, place-the-vial task). Docs-only; no source/test changes. * docs(groot): parameterize commands with env vars + fill LIBERO results - Introduce BASE_MODEL / DATASET_ID / REPO_ID / JOB_NAME / OUTPUT_DIR env vars in the training command and reuse OUTPUT_DIR + BASE_MODEL in the rollout cmd. - Fill the LIBERO benchmark table with GR00T-LeRobot success rates (Spatial 94%, Object 98%, Goal 93%, LIBERO 10/Long 90%; avg 93.75%), drop the OSS column and XX placeholders. LeRobot-focused. * docs(groot): drop export block, reference env vars directly Use $DATASET_ID / $BASE_MODEL / $REPO_ID / $OUTPUT_DIR / $JOB_NAME as bare placeholders in the commands without concrete export assignments. * docs(groot): keep BASE_MODEL export in training command * docs(groot): use literal HF repo IDs for dataset/policy repo_id Public-facing Hub references (--dataset.repo_id, --policy.repo_id) shown as concrete IDs; local-only values ($OUTPUT_DIR, $JOB_NAME) stay as placeholders. * docs(groot): add LIBERO training command example * docs(groot): remove LIBERO checkpoints subdirectory section * docs(groot): use $BASE_MODEL for base_model_path in LIBERO eval * docs(groot): drop hf download step from LIBERO eval, fix intro * docs(groot): restore suite checkpoint download intro sentence * docs(groot): remove checkpoint download note above LIBERO eval * docs(groot): update training and rollout commands with new parameters and dependencies * Add sample so101 training command * Remove sample so101 training command * docs(groot): remove optional Flash Attention setup instructions and update base model path for evaluation * docs(groot): update training command with image transformation parameters * docs(groot): add note on inference.queue_threshold value for stable inference * chore(style): pre-commit gr00t * docs(groot): update * chore(policies): minor details * fix(groot): license headers + test guards * chore(policies): fix tests * docs(groot): relative actions param doc * chore(policy): address some of the AI review items --------- Co-authored-by: Andrew Wrenn <awrenn@nvidia.com> Co-authored-by: Ryan Halabi <ryhalabi@nvidia.com> Co-authored-by: nv-sachdevkartik <ksachdev@nvidia.com> Co-authored-by: groot-validation <groot-validation@localhost> Co-authored-by: johnnynunez <johnnynuca14@gmail.com> Co-authored-by: lbenhorin <lbenhorin@nvidia.com>
2026-07-22 17:32:07 +00:00 · 2026-07-03 21:15:09 +02:00
parent e275ea3960
commit 708fa1d189
30 changed files with 8593 additions and 3474 deletions
@@ -1490,17 +1490,6 @@ wheels = [
    { url = "https://files.pythonhosted.org/packages/13/37/a065dc3bd6e49423a6532c642ca7378d3f467b1ef44c2800c937af7f9739/filelock-3.29.4-py3-none-any.whl", hash = "sha256:dac1648087d5115554850d113e7dd8c83ab2d38e3435dde2d4f163847e57b767", size = 42757, upload-time = "2026-06-13T16:11:59.582Z" },
 ]

-[[package]]
-name = "flash-attn"
-version = "2.8.3.post1"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "einops", marker = "platform_machine != 'arm64' or sys_platform != 'darwin'" },
-    { name = "torch", version = "2.11.0", source = { registry = "https://pypi.org/simple" }, marker = "(platform_machine != 'arm64' and sys_platform == 'darwin') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
-    { name = "torch", version = "2.11.0+cu128", source = { registry = "https://download.pytorch.org/whl/cu128" }, marker = "sys_platform == 'linux'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/01/7a/92a46e7cd6bbb4d7b2855a457c3b855df54a97af5656d98fc92e58e61065/flash_attn-2.8.3.post1.tar.gz", hash = "sha256:55d5103ed846da8b56e0797acf4bde07dee4b1c7e8907fcfc6699c203030c348", size = 8451657, upload-time = "2026-06-11T04:35:21.318Z" }
-
 [[package]]
 name = "flatbuffers"
 version = "25.12.19"
@@ -2825,8 +2814,10 @@ all = [
    { name = "contourpy" },
    { name = "datasets" },
    { name = "debugpy" },
+    { name = "decord", marker = "platform_machine == 'AMD64' or platform_machine == 'x86_64'" },
    { name = "deepdiff" },
    { name = "diffusers" },
+    { name = "dm-tree" },
    { name = "dynamixel-sdk" },
    { name = "faker" },
    { name = "fastapi" },
@@ -2875,6 +2866,7 @@ all = [
    { name = "scikit-image" },
    { name = "scipy" },
    { name = "teleop" },
+    { name = "timm" },
    { name = "torchcodec", marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine == 'AMD64' and sys_platform == 'linux') or (platform_machine == 'aarch64' and sys_platform == 'linux') or (platform_machine == 'arm64' and sys_platform == 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux') or sys_platform == 'win32'" },
    { name = "torchdiffeq" },
    { name = "transformers" },
@@ -2992,13 +2984,17 @@ gamepad = [
    { name = "pygame" },
 ]
 groot = [
+    { name = "av" },
+    { name = "datasets" },
    { name = "decord", marker = "platform_machine == 'AMD64' or platform_machine == 'x86_64'" },
    { name = "diffusers" },
    { name = "dm-tree" },
-    { name = "flash-attn", marker = "sys_platform != 'darwin'" },
-    { name = "ninja" },
+    { name = "jsonlines" },
+    { name = "pandas" },
    { name = "peft" },
+    { name = "pyarrow" },
    { name = "timm" },
+    { name = "torchcodec", marker = "(platform_machine == 'arm64' and sys_platform == 'darwin') or (platform_machine == 'AMD64' and sys_platform == 'linux') or (platform_machine == 'aarch64' and sys_platform == 'linux') or (platform_machine == 'arm64' and sys_platform == 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux') or sys_platform == 'win32'" },
    { name = "transformers" },
 ]
 grpcio-dep = [
@@ -3254,7 +3250,6 @@ requires-dist = [
    { name = "faker", marker = "extra == 'sarm'", specifier = ">=33.0.0,<35.0.0" },
    { name = "fastapi", marker = "extra == 'phone'", specifier = "<1.0" },
    { name = "feetech-servo-sdk", marker = "extra == 'feetech'", specifier = ">=1.0.0,<2.0.0" },
-    { name = "flash-attn", marker = "sys_platform != 'darwin' and extra == 'groot'", specifier = ">=2.5.9,<3.0.0" },
    { name = "foxglove-sdk", marker = "extra == 'viz'", specifier = ">=0.25.1,<0.26.0" },
    { name = "grpcio", marker = "extra == 'grpcio-dep'", specifier = ">=1.73.1,<2.0.0" },
    { name = "grpcio", marker = "extra == 'reachy2'", specifier = "<=1.73.1" },
@@ -3286,6 +3281,7 @@ requires-dist = [
    { name = "lerobot", extras = ["dataset"], marker = "extra == 'annotations'" },
    { name = "lerobot", extras = ["dataset"], marker = "extra == 'core-scripts'" },
    { name = "lerobot", extras = ["dataset"], marker = "extra == 'dataset-viz'" },
+    { name = "lerobot", extras = ["dataset"], marker = "extra == 'groot'" },
    { name = "lerobot", extras = ["dataset"], marker = "extra == 'hilserl'" },
    { name = "lerobot", extras = ["dataset"], marker = "extra == 'libero'" },
    { name = "lerobot", extras = ["dataset"], marker = "extra == 'metaworld'" },
@@ -3308,6 +3304,7 @@ requires-dist = [
    { name = "lerobot", extras = ["feetech"], marker = "extra == 'hopejr'" },
    { name = "lerobot", extras = ["feetech"], marker = "extra == 'lekiwi'" },
    { name = "lerobot", extras = ["gamepad"], marker = "extra == 'all'" },
+    { name = "lerobot", extras = ["groot"], marker = "extra == 'all'" },
    { name = "lerobot", extras = ["grpcio-dep"], marker = "extra == 'async'" },
    { name = "lerobot", extras = ["grpcio-dep"], marker = "extra == 'dev'" },
    { name = "lerobot", extras = ["grpcio-dep"], marker = "extra == 'hilserl'" },
@@ -3404,7 +3401,6 @@ requires-dist = [
    { name = "motorbridge", marker = "extra == 'motorbridge-dep'", specifier = ">=0.3.2,<0.4.0" },
    { name = "motorbridge-smart-servo", marker = "extra == 'motorbridge-smart-servo-dep'", specifier = ">=0.0.4,<0.1.0" },
    { name = "mypy", marker = "extra == 'dev'", specifier = ">=1.19.1" },
-    { name = "ninja", marker = "extra == 'groot'", specifier = ">=1.11.1,<2.0.0" },
    { name = "num2words", marker = "extra == 'smolvla'", specifier = ">=0.5.14,<0.6.0" },
    { name = "numpy", specifier = ">=2.0.0,<2.3.0" },
    { name = "onnx", marker = "extra == 'unitree-g1'", specifier = ">=1.16.0,<2.0.0" },
@@ -4194,32 +4190,6 @@ wheels = [
    { url = "https://files.pythonhosted.org/packages/9e/c9/b2622292ea83fbb4ec318f5b9ab867d0a28ab43c5717bb85b0a5f6b3b0a4/networkx-3.6.1-py3-none-any.whl", hash = "sha256:d47fbf302e7d9cbbb9e2555a0d267983d2aa476bac30e90dfbe5669bd57f3762", size = 2068504, upload-time = "2025-12-08T17:02:38.159Z" },
 ]

-[[package]]
-name = "ninja"
-version = "1.13.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/43/73/79a0b22fc731989c708068427579e840a6cf4e937fe7ae5c5d0b7356ac22/ninja-1.13.0.tar.gz", hash = "sha256:4a40ce995ded54d9dc24f8ea37ff3bf62ad192b547f6c7126e7e25045e76f978", size = 242558, upload-time = "2025-08-11T15:10:19.421Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3c/74/d02409ed2aa865e051b7edda22ad416a39d81a84980f544f8de717cab133/ninja-1.13.0-py3-none-macosx_10_9_universal2.whl", hash = "sha256:fa2a8bfc62e31b08f83127d1613d10821775a0eb334197154c4d6067b7068ff1", size = 310125, upload-time = "2025-08-11T15:09:50.971Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/de/6e1cd6b84b412ac1ef327b76f0641aeb5dcc01e9d3f9eee0286d0c34fd93/ninja-1.13.0-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3d00c692fb717fd511abeb44b8c5d00340c36938c12d6538ba989fe764e79630", size = 177467, upload-time = "2025-08-11T15:09:52.767Z" },
-    { url = "https://files.pythonhosted.org/packages/c8/83/49320fb6e58ae3c079381e333575fdbcf1cca3506ee160a2dcce775046fa/ninja-1.13.0-py3-none-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:be7f478ff9f96a128b599a964fc60a6a87b9fa332ee1bd44fa243ac88d50291c", size = 187834, upload-time = "2025-08-11T15:09:54.115Z" },
-    { url = "https://files.pythonhosted.org/packages/56/c7/ba22748fb59f7f896b609cd3e568d28a0a367a6d953c24c461fe04fc4433/ninja-1.13.0-py3-none-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:60056592cf495e9a6a4bea3cd178903056ecb0943e4de45a2ea825edb6dc8d3e", size = 202736, upload-time = "2025-08-11T15:09:55.745Z" },
-    { url = "https://files.pythonhosted.org/packages/79/22/d1de07632b78ac8e6b785f41fa9aad7a978ec8c0a1bf15772def36d77aac/ninja-1.13.0-py3-none-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:1c97223cdda0417f414bf864cfb73b72d8777e57ebb279c5f6de368de0062988", size = 179034, upload-time = "2025-08-11T15:09:57.394Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/de/0e6edf44d6a04dabd0318a519125ed0415ce437ad5a1ec9b9be03d9048cf/ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fb46acf6b93b8dd0322adc3a4945452a4e774b75b91293bafcc7b7f8e6517dfa", size = 180716, upload-time = "2025-08-11T15:09:58.696Z" },
-    { url = "https://files.pythonhosted.org/packages/54/28/938b562f9057aaa4d6bfbeaa05e81899a47aebb3ba6751e36c027a7f5ff7/ninja-1.13.0-py3-none-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:4be9c1b082d244b1ad7ef41eb8ab088aae8c109a9f3f0b3e56a252d3e00f42c1", size = 146843, upload-time = "2025-08-11T15:10:00.046Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/fb/d06a3838de4f8ab866e44ee52a797b5491df823901c54943b2adb0389fbb/ninja-1.13.0-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:6739d3352073341ad284246f81339a384eec091d9851a886dfa5b00a6d48b3e2", size = 154402, upload-time = "2025-08-11T15:10:01.657Z" },
-    { url = "https://files.pythonhosted.org/packages/31/bf/0d7808af695ceddc763cf251b84a9892cd7f51622dc8b4c89d5012779f06/ninja-1.13.0-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:11be2d22027bde06f14c343f01d31446747dbb51e72d00decca2eb99be911e2f", size = 552388, upload-time = "2025-08-11T15:10:03.349Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/70/c99d0c2c809f992752453cce312848abb3b1607e56d4cd1b6cded317351a/ninja-1.13.0-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:aa45b4037b313c2f698bc13306239b8b93b4680eb47e287773156ac9e9304714", size = 472501, upload-time = "2025-08-11T15:10:04.735Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/43/c217b1153f0e499652f5e0766da8523ce3480f0a951039c7af115e224d55/ninja-1.13.0-py3-none-musllinux_1_2_i686.whl", hash = "sha256:5f8e1e8a1a30835eeb51db05cf5a67151ad37542f5a4af2a438e9490915e5b72", size = 638280, upload-time = "2025-08-11T15:10:06.512Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/45/9151bba2c8d0ae2b6260f71696330590de5850e5574b7b5694dce6023e20/ninja-1.13.0-py3-none-musllinux_1_2_ppc64le.whl", hash = "sha256:3d7d7779d12cb20c6d054c61b702139fd23a7a964ec8f2c823f1ab1b084150db", size = 642420, upload-time = "2025-08-11T15:10:08.35Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/fb/95752eb635bb8ad27d101d71bef15bc63049de23f299e312878fc21cb2da/ninja-1.13.0-py3-none-musllinux_1_2_riscv64.whl", hash = "sha256:d741a5e6754e0bda767e3274a0f0deeef4807f1fec6c0d7921a0244018926ae5", size = 585106, upload-time = "2025-08-11T15:10:09.818Z" },
-    { url = "https://files.pythonhosted.org/packages/c1/31/aa56a1a286703800c0cbe39fb4e82811c277772dc8cd084f442dd8e2938a/ninja-1.13.0-py3-none-musllinux_1_2_s390x.whl", hash = "sha256:e8bad11f8a00b64137e9b315b137d8bb6cbf3086fbdc43bf1f90fd33324d2e96", size = 707138, upload-time = "2025-08-11T15:10:11.366Z" },
-    { url = "https://files.pythonhosted.org/packages/34/6f/5f5a54a1041af945130abdb2b8529cbef0cdcbbf9bcf3f4195378319d29a/ninja-1.13.0-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:b4f2a072db3c0f944c32793e91532d8948d20d9ab83da9c0c7c15b5768072200", size = 581758, upload-time = "2025-08-11T15:10:13.295Z" },
-    { url = "https://files.pythonhosted.org/packages/95/97/51359c77527d45943fe7a94d00a3843b81162e6c4244b3579fe8fc54cb9c/ninja-1.13.0-py3-none-win32.whl", hash = "sha256:8cfbb80b4a53456ae8a39f90ae3d7a2129f45ea164f43fadfa15dc38c4aef1c9", size = 267201, upload-time = "2025-08-11T15:10:15.158Z" },
-    { url = "https://files.pythonhosted.org/packages/29/45/c0adfbfb0b5895aa18cec400c535b4f7ff3e52536e0403602fc1a23f7de9/ninja-1.13.0-py3-none-win_amd64.whl", hash = "sha256:fb8ee8719f8af47fed145cced4a85f0755dd55d45b2bddaf7431fa89803c5f3e", size = 309975, upload-time = "2025-08-11T15:10:16.697Z" },
-    { url = "https://files.pythonhosted.org/packages/df/93/a7b983643d1253bb223234b5b226e69de6cda02b76cdca7770f684b795f5/ninja-1.13.0-py3-none-win_arm64.whl", hash = "sha256:3c0b40b1f0bba764644385319028650087b4c1b18cdfa6f45cb39a3669b81aa9", size = 290806, upload-time = "2025-08-11T15:10:18.018Z" },
-]
-
 [[package]]
 name = "nodeenv"
 version = "1.10.0"