OpenVLA (7B) and OpenVLA-OFT
Model idopenvla/openvla-7b
PricePrice not published
Checked29 September 2026
Where it stands
32newer robotics models since its release
- Praxis-130 Sep 2026
- Helix 2.517 Sep 2026
- UnifoLM-WLA-1.010 Sep 2026
- Skild S1 (Skild Brain)25 Aug 2026
- Generalist GEN-1.519 Aug 2026
The 5 newest are listed. All robotics models
- Weights
- Weights are public under a permissive licence (MIT).
How to call it
3 ways to use OpenVLA (7B) and OpenVLA-OFT, with the model id each provider expects.
| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Hugging Face | openvla/openvla-7b |
huggingface.co | |
| Hugging Face (OFT fine-tunes) | moojink/openvla-7b-oft-finetuned-libero-spatial |
huggingface.co | |
| GitHub | openvla.github.io github.com |
Pricing
Price not published.
No price source linked. Checked 29 September 2026.
Specs
- Input
- Images and text
- Output
- Robot actions
- Context window
- Not published
- Max output
- Not published
- Open weights
- Yes
- Licence
- MIT
- Released
- 13 June 2024
- Status
- Legacy
- Type
- Robotics model
What stands out
What its maker marketed at launch, and what people found later, each with its source.
Open 7B generalist VLA beating a 55B closed model
Llama 2 7B backbone with fused DINOv2 + SigLIP vision, trained on ~970k Open X-Embodiment episodes; outperformed RT-2-X (55B) by 16.5% absolute success over 29 tasks with 7x fewer parameters, and fine-tunes with LoRA on consumer GPUs.
OFT fine-tuning recipe (Feb 2025)
Found after launch
OpenVLA-OFT (parallel decoding, action chunking, continuous actions, L1 loss) raised LIBERO average success from 76.5% to 97.1% and action throughput 26x; on bimanual ALOHA it beat pi0 and RDT-1B by up to 15% absolute.
Notes
The most-downloaded open VLA checkpoint on HF (500k+ downloads at check time); widely used as a research baseline. Superseded in capability by pi0-family and newer open VLAs but still a standard reference. Release day: arXiv 2406.09246 v1 dated 2024-06-13 (HF repo created 2024-06-10).