NORA
Robotics model3.76B-parameter Vision-Language-Action model fine-tuned from Qwen2.5-VL-3B on Open X-Embodiment with the FAST action tokenizer, offering single-step (NORA) and 5-step chunking (NORA-Long) checkpoints plus RLDS and LeRobot fine-tuning scripts. [1] [2] [3] [4]
Setup & usage →CheckpointsModel hub / homepageGetting started
- For
- Robotics researchers exploring compact Qwen2.5-VL + FAST VLA policies; Practitioners fine-tuning 7-DoF manipulation policies on RLDS or LeRobot datasets [1] [2]
- Embodiments
- WidowX (BridgeData V2 real robot and OpenVLA eval server), Franka Panda (LIBERO simulation finetuned checkpoints),… [1] [2] [3]
- Source rights
- Named licenses are documented by part. Review licenses
Source researched · Not tested by us
Sources checked · Research details
Licenses & related hardware
- Ecosystem
- LeRobot
Overview
NORA (declare-lab/nora, arXiv:2504.19854) is a 3,758,262,272-parameter (~3.76B) Vision-Language-Action model developed by DeCLaRe Lab by fine-tuning Qwen2.5-VL-3B on Open X-Embodiment manipulation episodes. Rather than training a separate diffusion action head, NORA predicts discrete action tokens (vocabulary IDs 151665–153712) decoded at inference time by the physical-intelligence/fast FAST tokenizer into normalized 7-DoF end-effector deltas (x, y, z, roll, pitch, yaw, gripper). [1] [2] [4]
Model card summary
- Architecture
Qwen2.5-VL-3B (Qwen2_5_VLForConditionalGeneration, 36 layers, hidden_size 2048) fine-tuned autoregressively to emit discrete robot action tokens (token IDs 151665–153712) decoded via the physical-intelligence/fast FAST tokenizer.
- Parameter count
3,758,262,272 (~3.76B BF16 parameters per HF safetensors metadata)
Verified identical safetensors parameter count for both declare-lab/nora and declare-lab/nora-long.
- Input modalities
- rgblanguage
Single-camera workspace RGB image and English language instruction.
- Output modalities
- ee-pose-deltagripper-command
Normalized 7-DoF end-effector deltas (x, y, z, roll, pitch, yaw, gripper in [0, 1]).
- Action space
FAST-tokenized 7-DoF normalized end-effector deltas (6-DoF pose delta + gripper in [0, 1]); single-step in declare-lab/nora and 5-step action horizon in declare-lab/nora-long, unnormalized per dataset statistics.
- Trained / supported embodiments
- WidowX (BridgeData V2 real robot and OpenVLA eval server)
- Franka Panda (LIBERO simulation finetuned checkpoints)
- 7-DoF EEF-delta manipulators via RLDS or LeRobot fine-tuning
- Training datasets
- Open X-Embodiment (OXE) RLDS robot manipulation mixture
- LIBERO task suites (for separate LIBERO-finetuned checkpoints in the NORA HF collection)
Checkpoints
| Checkpoint | Format | Size | Access | Pinned revision |
|---|---|---|---|---|
| declare-lab/nora | safetensors | 7 GB | open | 3f373eff454d |
| declare-lab/nora-long | safetensors | 7 GB | open | 1fd12607a728 |
Two primary ungated base checkpoints (7.0 GB BF16 safetensors each); the NORA HF collection also hosts 8 LIBERO-finetuned checkpoints (4 for nora, 4 for nora-long) and successor declare-lab/nora-1.5.
Is this for you?
- Use cases
- Zero-shot 7-DoF instruction following on WidowX BridgeData V2 setups
- Fine-tuning on custom RLDS or LeRobot 7-DoF manipulation datasets with Accelerate
- Action-chunking manipulation in LIBERO simulation via NORA-Long
- Known limitations
- Pretrained on a 7-DoF end-effector delta action space (6-DoF + gripper); upstream warns fine-tuning on other action spaces may not work well
- GitHub repository declare-lab/nora has no root LICENSE file and Hugging Face model card YAML omits the license field, even though model card prose states MIT
- NORA-Long (horizon 5) performs worse on WidowX Bridge tasks despite stronger LIBERO simulation results
- Gripper convention is normalized to [0, 1] (0=close, 1=open), requiring the invert_grippler_action flag during training and normalize_gripper_action plus invert_gripper_action at inference for [-1, 1] environments such as LIBERO
- Safety and limitations
- Inference loads the external physical-intelligence/fast tokenizer with trust_remote_code=True from Hugging Face.
- No root LICENSE file is present in declare-lab/nora on GitHub (MIT is stated only in Hugging Face model card prose), creating licensing ambiguity for formal compliance audits.
- Pretrained exclusively on 7-DoF end-effector delta actions without proprioception or robot-side safety limits.
Compute & fine-tuning
- Inference compute
- Documented hardware: CUDA GPU or CPU fallback (exact GPU model and VRAM not stated in README)
- Precision:
bf16 - Edge / consumer capable: No
7.0 GB BF16 safetensors checkpoint; exact minimum VRAM and latency_ms are not documented in README or model card.
- Fine-tuning
- Supported: Yes
- Methods: Multi-GPU RLDS fine-tuning via Accelerate (training/train.py); LeRobot dataset fine-tuning via Accelerate (lerobot_training/lerobot_training.py)
- Recipe: Upstream fine-tuning documentation
Changing action horizon requires editing future_action_window_size in training/datasets/datasets.py.
- Host computer
Single CUDA GPU (or CPU fallback in inference/nora.py) for bfloat16 inference on the 7.0 GB checkpoint; multi-GPU Accelerate cluster documented for RLDS/LeRobot pretraining and fine-tuning (exact minimum VRAM not stated upstream).
Exact GPU VRAM figures are not specified in the GitHub README or model cards.
- Operating system
- Linux
Standard Linux conda, CUDA, and Accelerate workflow documented.
[1] · Editorial assessment
- Dependencies
- Python 3.10
- PyTorch / torchvision (>= 0.21.0 for LeRobot fine-tuning)
- transformers (4.50.0 recorded in config.json)
- accelerate (for multi-GPU training)
- physical-intelligence/fast (external Hugging Face FAST action tokenizer loaded with trust_remote_code)
- lerobot (optional, for LeRobot dataset fine-tuning)
- Skills
- Python 3.10 and conda
- PyTorch, Hugging Face Transformers, and Accelerate
- RLDS or LeRobot dataset preparation
Inferred from documented conda, accelerate, RLDS, and LeRobot training workflows.
[1] · Editorial assessment
Setup & usage
Open the upstream getting-started guide [1] · Model card / documentation [1]
- Clone declare-lab/nora and create a Python 3.10 conda environment
- Install inference/requirements.txt (or training/requirements.txt for RLDS, lerobot_training/lerobot_requirements.txt for LeRobot datasets)
- Instantiate Nora(device='cuda') from inference.nora (downloads declare-lab/nora and physical-intelligence/fast tokenizer)
- Call nora.inference(image, instruction, unnorm_key='bridge_orig') or pass a LeRobot Unnormalize transform
Upstream-reported benchmarks
Upstream-reported only: Open Source Robotics has not run training, inference, or hardware evaluations for these benchmarks.
| Benchmark | Upstream-reported score | Scope & caveats |
|---|---|---|
| NORA linked paper evaluations (arXiv:2504.19854; WidowX BridgeData V2 and LIBERO) | No numeric benchmark tables in README or model cards (model cards note NORA performs better on WidowX Bridge and NORA-Long performs better on LIBERO; see linked arXiv:2504.19854) | Qualitative comparison is upstream-reported in model cards; OSR has not inspected arXiv:2504.19854 tables or run evaluations. |
Project notes
Base checkpoints, horizon tradeoff, and fine-tuning routes
Two primary ungated BF16 safetensors checkpoints are hosted on Hugging Face at 7.0 GB each: declare-lab/nora (single-step prediction) and declare-lab/nora-long (pretrained with a 5-step action horizon), alongside eight LIBERO-finetuned checkpoints and a newer declare-lab/nora-1.5 release in the same collection. Upstream notes a clear empirical tradeoff: nora-long performs worse than single-step nora on WidowX Bridge tasks while performing better in LIBERO simulation. Note that the inspected inference/nora.py wrapper hard-codes the FAST tokenizer's time_horizon to 1, so users running declare-lab/nora-long should verify how five-step decoding is configured before using that wrapper for five-step predictions. Both RLDS (OpenVLA-style) and LeRobot dataset fine-tuning are supported via Hugging Face Accelerate: RLDS training runs training/train.py with accelerate launch and a cluster-matched Accelerate config, while LeRobot training runs lerobot_training/lerobot_training.py in a separate environment requiring torchvision>=0.21.0; in both routes, dataset paths and hyperparameters are configured in TrainingConfig, and changing the action horizon requires modifying future_action_window_size in training/datasets/datasets.py. Two action-space conventions matter when adapting NORA: pretraining uses 7-DoF end-effector deltas exclusively (upstream cautions that other action spaces may not fine-tune well) and pose outputs require per-robot, per-dataset unnormalization (unnorm_key or a LeRobot Unnormalize transform). In addition, NORA normalizes gripper actions to [0, 1] (0 = close, 1 = open): for environments using [-1, 1] conventions such as LIBERO (-1 = open, +1 = close), upstream documents the invert_grippler_action flag in TrainingConfig during training and the inference helpers normalize_gripper_action(..., binarize=True) followed by invert_gripper_action(...) to map predictions back to [-1, 1] with inverted sign. [1] [2] [3] [4]
Compute gaps and upstream-reported evaluations
The lightweight inference/nora.py wrapper loads declare-lab/nora in bfloat16 on CUDA (with CPU fallback) and fetches physical-intelligence/fast with trust_remote_code=True. However, neither the GitHub README nor the Hugging Face model cards report minimum GPU VRAM for inference or fine-tuning, per-step latency, or closed-loop control frequency (Hz). Likewise, the supplied README and model cards do not include numeric benchmark tables for WidowX BridgeData V2 or LIBERO, linking instead to arXiv:2504.19854 for paper results; Open Source Robotics has not inspected the paper tables or run training, fine-tuning, inference, or robot evaluations for this profile. [1] [2] [3] [4]
What we checked
None. Prepared from public upstream sources and Hugging Face API metadata without hands-on training, fine-tuning, or inference.
Not yet established
- Documented closed-loop control frequency (Hz)
Updates
- Upstream last updated
- Jan 9, 2026 [1]
- Profile updated
- Nov 21, 2026
- Sources last checked
- Nov 21, 2026
- Next review
- Feb 19, 2027
- Reviewed revision
- declare-lab/nora@6b18c23d7875052e03fba4f8c2f32bd6a8a5c4a9; HF declare-lab/nora@3f373eff454d299bad4aa0d845a5c0db4fb0e0bc · combined hardware and software repository
Separate license matrix
Open-weights 3.76B Qwen2.5-VL + FAST VLA with ungated safetensors checkpoints and RLDS/LeRobot training code; classified as partial openness because declare-lab/nora lacks a root LICENSE file and HF YAML metadata omits a machine-readable license tag (MIT is stated only in model card prose).
| Layer | Availability / license |
|---|---|
| Code | available · MIT (declared in HF model card prose; no root LICENSE file in GitHub repo) GitHub API reports license: null and root directory contains no LICENSE file; HF model cards explicitly state 'All Nora checkpoints, as well as our training codebase are released under an MIT License.' [1] [2] |
| Model weights | available · MIT (declared in HF model card prose; omitted from HF YAML frontmatter) HF API cardData.license is null because YAML frontmatter omits license:, though model card body lists License: MIT. [2] [3] |
| Training datasets | available · Mixed (Open X-Embodiment RLDS mixture; per-dataset upstream licenses apply) [1] [2] |
| Documentation | available · MIT (declared in HF model card prose; no root LICENSE file in GitHub repo) [1] [2] |
Licensing hygiene gap between GitHub and Hugging Face
A notable compliance caveat separates NORA's prose claims from its repository metadata: both the declare-lab/nora and declare-lab/nora-long model card bodies state that all NORA checkpoints and the training codebase are released under an MIT License, yet the declare-lab/nora GitHub repository has no root LICENSE file (license: null via GitHub API) and the Hugging Face YAML frontmatter omits a license: tag. Teams with strict license-scanning pipelines should verify license intent directly with the maintainers and audit individual Open X-Embodiment dataset licenses before commercial deployment. [1] [2] [3]
References
Each source links to the exact revision we read. “Used for” shows where this page relies on it.
- S001 declare-lab/nora GitHub repository metadata and README.md
No root LICENSE file present; documents inference, RLDS and LeRobot Accelerate fine-tuning, 7-DoF action space, gripper [0, 1] normalization, and arXiv:2504.19854.Revision, date & where cited
DeCLaRe Lab (GitHub) · github.com · revision
6b18c23· GitHub API, root tree, and README.md at commit 6b18c23d7875 (2026-01-09) · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Is this for you?, Compute & fine-tuning and 6 more - S002 Hugging Face model card and API metadata: declare-lab/nora
3,758,262,272 BF16 parameters, 7.0 GB ungated safetensors, fine-tuned from Qwen 2.5 VL-3B on Open X-Embodiment; model card body declares MIT license for checkpoints and training codebase (YAML license omitted).Revision, date & where cited
declare-lab (Hugging Face) · huggingface.co · HF API & README.md at revision 3f373eff454d (lastModified 2025-08-27) · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Checkpoints, Is this for you? and 5 more - S003 Hugging Face model card and API metadata: declare-lab/nora-long
3,758,262,272 BF16 parameters, 7.0 GB ungated safetensors, pretrained with action horizon 5; notes worse WidowX performance but superior LIBERO simulation performance.Revision, date & where cited
declare-lab (Hugging Face) · huggingface.co · HF API & README.md at revision 1fd12607a728 (lastModified 2025-08-27) · Accessed Nov 21, 2026
Used for: Model card summary, Checkpoints, Is this for you?, Upstream-reported benchmarks and 4 more - S004 declare-lab/nora inference/nora.py and HF config.json
Qwen2_5_VLForConditionalGeneration (36 layers, hidden_size 2048), action token range 151665–153712, loads physical-intelligence/fast tokenizer with trust_remote_code, defaults to bfloat16 on CUDA or CPU.Revision, date & where cited
DeCLaRe Lab (GitHub) · github.com · revision
6b18c23· inference/nora.py at commit 6b18c23d7875 & config.json at revision 3f373eff454d · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Is this for you?, Compute & fine-tuning and 3 more
Correction guidance · Discussion is not enabled · Profile format 0.8.0-draft.1
Change history
2026-11-21 — Initial 0.8.0-draft.1 model profile prepared from declare-lab/nora@6b18c23d7875 and HF declare-lab/nora + nora-long checkpoints.