SmolVLA
Robotics modelCompact 450M-parameter flow-matching Vision-Language-Action base model built on SmolVLM2-500M-Video-Instruct for affordable single-GPU fine-tuning and consumer hardware deployment via LeRobot, with Apache-2.0 weights and pretraining reproduction kit (VLAb). [1] [2] [3]
Setup & usage →CheckpointsModel hub / homepageSource codeGetting started
- For
- Low-cost robot builders (SO-100/SO-101, LeRobot users); Robotics researchers; Single-GPU VLA practitioners [1] [3]
- Parameters
- 450,046,176 (~450M [1]
- Embodiments
- so-101, SO-100, LeRobot community-compatible arms (padded up to 32-D state/action), … [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
SmolVLA (lerobot/smolvla_base) is a 450,046,176-parameter (~450M) flow-matching Vision-Language-Action base model built on HuggingFaceTB/SmolVLM2-500M-Video-Instruct. It is packaged for consumer-hardware fine-tuning and deployment through lerobot[smolvla], with a separate multi-GPU pretraining reproduction repository (huggingface/vlab). Note that checkpoints produced by huggingface/vlab may be incompatible with newer LeRobot normalization formats and may require the documented normalization migration before LeRobot fine-tuning or inference; this caveat applies specifically to VLAb-produced checkpoints. [1] [2] [3]
Model card summary
- Architecture
SmolVLM2-500M-Video-Instruct vision-language backbone paired with a compact flow-matching action expert (10 denoising steps by default).
- Base model
HuggingFaceTB/SmolVLM2-500M-Video-Instruct
- Parameter count
450,046,176 (~450M; 446,772,624 BF16 + 3,273,552 F32 per safetensors metadata)
Exact tensor count verified via Hugging Face API safetensors metadata.
- Input modalities
- rgbproprioceptionlanguage
Up to 3 RGB cameras (resized with padding to 512x512; default input shape 3x256x256), proprioceptive state (default 6-D, padded up to max_state_dim: 32), optional English language instruction.
- Output modalities
- joint-positionsgripper-command
Continuous action chunks (default shape 6, padded up to max_action_dim: 32) for joint/gripper control.
- Action space
Continuous flow-matching action chunks with chunk_size: 50, n_action_steps: 50, default 6-DoF action vector padded up to max_action_dim: 32, and 10 flow-matching integration steps (num_steps: 10).
Verified directly from config.json at revision 5e8d12a6e297.
- Trained / supported embodiments
- so-101
- SO-100
- LeRobot community-compatible arms (padded up to 32-D state/action)
- LIBERO simulation (Franka)
Quickstart demonstrates lerobot/libero; designed for affordable LeRobot arms.
- Training datasets
- LeRobot community datasets on Hugging Face Hub (multi-dataset mixture documented in huggingface/vlab)
Trained on community-driven Hugging Face robotics datasets; exact dataset list is maintained in VLAb reproduction configs.
Checkpoints
| Checkpoint | Format | Size | Access | Pinned revision |
|---|---|---|---|---|
| lerobot/smolvla_base | safetensors | 0.85 GB | open | 5e8d12a6e297 |
Ungated safetensors checkpoint (0.85 GB) verified via Hugging Face API.
Is this for you?
- Use cases
- Fine-tuning compact VLA policies on custom LeRobot datasets
- Consumer-GPU robot manipulation inference
- Multi-GPU VLA pretraining reproduction via VLAb
- Known limitations
- Intended as a base model for task-specific fine-tuning rather than zero-shot deployment on arbitrary new setups
- Runtime inference and fine-tuning depend on the separate huggingface/lerobot package
Compute & fine-tuning
- Inference compute
- Documented hardware: Consumer single GPU / CPU (per model card)
- Precision:
bf16 - Edge / consumer capable: Yes
0.85 GB BF16/F32 safetensors footprint supports consumer hardware and CPU/GPU inference; exact minimum VRAM and latency_ms not stated in model card.
[1] · Editorial assessment
- Fine-tuning
- Supported: Yes
- Methods: full fine-tune via LeRobot; multi-GPU pretraining reproduction via VLAb
- Recipe: Upstream fine-tuning documentation
Pinned config.json (S002) sets freeze_vision_encoder: true, train_expert_only: true, and train_state_proj: true by default; VLAb-produced checkpoints may require normalization migration before LeRobot fine-tuning or inference (S003).
- Host computer
Designed to be trainable on a single GPU and deployable on consumer hardware (CUDA or CPU supported in quickstart).
- Operating system
- Linux
Standard Python/PyTorch/conda Linux workflow documented.
- Dependencies
- Python
- PyTorch
- lerobot[smolvla]
- HuggingFaceTB/SmolVLM2-500M-Video-Instruct
- Accelerate (for VLAb pretraining)
- Skills
- Python
- PyTorch
- LeRobot CLI/API
Inferred from documented Python/PyTorch/CLI setup and training dependencies.
[1] · Editorial assessment
Setup & usage
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 |
|---|---|---|
| SmolVLA paper evaluations (arXiv:2506.01844; LIBERO, Meta-World, real-robot tasks) | See arXiv:2506.01844 (model card states performance matching much larger VLAs) | Upstream-reported benchmark claims; exact tables in arXiv:2506.01844 not independently verified by OSR. |
Project notes
Compact architecture and padded 32-D action space
At the pinned Hub revision (5e8d12a6e297), config.json specifies continuous flow-matching action chunks with chunk_size: 50, n_action_steps: 50, and num_steps: 10 denoising integration steps. Inputs accept up to three RGB camera streams (resized with padding to 512x512), proprioceptive state (default 6-D, padded up to max_state_dim: 32), and optional English language instructions, outputting continuous joint/gripper actions padded up to max_action_dim: 32. For fine-tuning, config.json sets freeze_vision_encoder: true, train_expert_only: true, and train_state_proj: true by default—meaning the default configuration updates the action expert and state projection rather than every parameter in the model. The checkpoint is intended as a base model for task-specific fine-tuning on target robot demonstrations rather than zero-shot control on arbitrary hardware. [1] [2]
Deployment footprint and upstream-reported benchmarks
The ungated safetensors checkpoint totals 0.85 GB (BF16/F32), and the model card documents single-GPU training as well as consumer GPU or CPU inference via LeRobot. Exact minimum VRAM and closed-loop control frequency are not pinned as single scalars in config.json and depend on the host and asynchronous inference setup. Benchmark claims in arXiv:2506.01844 (covering LIBERO, Meta-World, and real-robot tasks) are upstream-reported; Open Source Robotics has not run training, fine-tuning, inference, or robot evaluations for this profile. [1] [2]
What we checked
None. Prepared from public upstream sources without hands-on hardware execution.
Not yet established
- Documented closed-loop control frequency (Hz)
Updates
- Upstream last updated
- Oct 6, 2026 [1]
- Latest release
- Oct 6, 2026 [1]
- Profile updated
- Nov 21, 2026
- Sources last checked
- Nov 21, 2026
- Next review
- May 21, 2027
- Reviewed revision
- HF lerobot/smolvla_base@5e8d12a6e2975b0e5e5fce7c8caf47c371d257b6; huggingface/vlab@10558f6f958902c1b5ff5eed76ff5766fab6f64b · combined hardware and software repository
Separate license matrix
Standalone 450M open-weights flow-matching VLA model with verified Apache-2.0 weights and Apache-2.0 pretraining reproduction kit (VLAb); profiled narrowly as a model artifact while linking to existing lerobot runtime.
| Layer | Availability / license |
|---|---|
| Code | available · Apache-2.0 GitHub API reports NOASSERTION due to formatting, but root LICENSE file at commit 10558f6f9589 is verbatim Apache License Version 2.0. [3] [4] |
| Model weights | available · Apache-2.0 Model card YAML declares license: apache-2.0; base model SmolVLM2-500M-Video-Instruct is also Apache-2.0. [1] |
| Training datasets | available · Mixed (community-contributed Hugging Face LeRobot datasets; per-dataset licenses apply) Pretraining relies on multiple public community datasets on Hugging Face Hub with individual dataset licenses. [1] [3] |
| Documentation | available · Apache-2.0 [1] |
Separate code, weights, and community dataset rights
Both the lerobot/smolvla_base checkpoint and the huggingface/vlab pretraining repository carry Apache-2.0 licenses (verified directly in vlab's root LICENSE file and the model card YAML, alongside the Apache-2.0 SmolVLM2-500M-Video-Instruct base model). Pretraining draws on community-contributed LeRobot datasets hosted on Hugging Face Hub, where individual upstream dataset licenses and data quality vary by contributor and must be reviewed separately before commercial reuse. [1] [3] [4]
References
Each source links to the exact revision we read. “Used for” shows where this page relies on it.
- S001 Hugging Face model card and API metadata: lerobot/smolvla_base
450,046,176 parameters (446,772,624 BF16 + 3,273,552 F32), 0.85 GB safetensors, ungated, license: apache-2.0, base_model: HuggingFaceTB/SmolVLM2-500M-Video-Instruct, lastModified 2026-10-06.Revision, date & where cited
Hugging Face (lerobot) · huggingface.co · HF API & README.md at revision 5e8d12a6e297 · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Checkpoints, Is this for you? and 7 more - S002 lerobot/smolvla_base config.json at pinned revision
chunk_size: 50, n_action_steps: 50, max_state_dim: 32, max_action_dim: 32, 3 camera inputs resized to 512x512, num_steps: 10.Revision, date & where cited
Hugging Face (lerobot) · huggingface.co · config.json at revision 5e8d12a6e297 · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Compact architecture and padded 32-D action space, Deployment footprint and upstream-reported benchmarks - S003 huggingface/vlab GitHub repository metadata and README.md
Official multi-GPU and SLURM pretraining reproduction kit for SmolVLA derived from LeRobot.Revision, date & where cited
Hugging Face (GitHub) · github.com · revision
10558f6· GitHub API & README.md at commit 10558f6f9589 (2025-11-19) · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Is this for you?, Compute & fine-tuning and 2 more - S004 huggingface/vlab LICENSE at pinned commit
Revision, date & where cited
Hugging Face · github.com · revision
10558f6· LICENSE at commit 10558f6f9589 · Accessed Nov 21, 2026
Used for: Separate license matrix
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 draft prepared from HF revision 5e8d12a6e297 and vlab commit 10558f6f9589.