UniVLA
Robotics model7.54B-parameter generalist Vision-Language-Action model (RSS 2025) pretrained on Open X-Embodiment and Ego4D human videos via task-centric embodiment-agnostic latent actions, with Apache-2.0 code and weights. [1] [2] [3]
Setup & usage →CheckpointsModel hub / homepageGetting started
- For
- Robotics researchers; Cross-embodiment VLA developers; Video-to-action pretraining teams [4]
- Parameters
- 7,541,237,184 (~7.54B BF16 parameters per safetensors metadata) [2]
- Embodiments
- Franka Panda (LIBERO / CALVIN), WidowX 250 (BridgeV2 / SimplerEnv), Google Robot (SimplerEnv), … [4]
- Source rights
- Named licenses are documented by part. Review licenses
Source researched · Not tested by us
Sources checked · Research details
Licenses & related hardware
Overview
UniVLA (qwbu/univla-7b) is a 7,541,237,184-parameter (~7.54B) generalist Vision-Language-Action model (RSS 2025) trained in two stages: a task-centric Latent Action Model (LAM) paired with a 7B Prismatic VLM backbone (TRI-ML/prismatic-vlms/prism-dinosiglip-224px+7b, combining DINOv2 and SigLIP vision encoders with a 7B LLM). Pretraining draws on Open X-Embodiment robot demonstrations and Ego4D human egocentric videos. [1] [2] [4]
Model card summary
- Architecture
Two-stage task-centric Latent Action Model (LAM) paired with a 7B Prismatic VLM backbone (TRI-ML/prismatic-vlms prism-dinosiglip-224px+7b, OpenVLA-style DINOv2 + SigLIP vision encoders + 7B LLM) predicting discrete latent action tokens decoded into robot actions during post-training.
- Base model
TRI-ML/prismatic-vlms/prism-dinosiglip-224px+7b
Explicitly listed in UniVLA README Model Zoo table.
- Parameter count
7,541,237,184 (~7.54B BF16 parameters per safetensors metadata)
Exact tensor count verified via Hugging Face API safetensors metadata.
- Input modalities
- rgblanguage
RGB visual observations (224px DINOv2+SigLIP) and language task instructions.
- Output modalities
- ee-pose-deltagripper-commandwaypoints
Pretrained model outputs embodiment-agnostic latent action tokens; post-trained policies decode to manipulation actions (LIBERO/CALVIN/SimplerEnv/real-world) or navigation actions (Room2Room).
- Action space
Unified discrete task-centric latent action tokens during pretraining, decoded to embodiment-specific continuous manipulation or navigation action spaces via post-training action decoder.
- Trained / supported embodiments
- Franka Panda (LIBERO / CALVIN)
- WidowX 250 (BridgeV2 / SimplerEnv)
- Google Robot (SimplerEnv)
- Navigation agent (Room2Room)
- Real-world tabletop arm (per README demo videos)
- Training datasets
- open-x-embodiment
- bridgedata-v2
- Ego4D (human egocentric videos)
Full univla-7b pretrained on OpenX + Ego4D (manipulation, navigation, and human videos).
Checkpoints
| Checkpoint | Format | Size | Access | Pinned revision |
|---|---|---|---|---|
| qwbu/univla-7b (full OpenX + Ego4D pretrained) | safetensors | 14.05 GB | open | 968e2896809c |
Additional checkpoints (univla-latent-action-model, univla-7b-bridge-pt, univla-7b-human-pt, univla-libero) are linked in the README Model Zoo.
Is this for you?
- Use cases
- Pretraining VLAs from unlabeled human and cross-embodiment videos
- Post-training/fine-tuning on manipulation and navigation benchmarks (LIBERO, CALVIN, SimplerEnv, Room2Room)
- Real-robot manipulation transfer
- Known limitations
- Pretrained univla-7b plans in latent action space and requires post-training / action-decoder adaptation for a specific robot embodiment
- Backbone derives from Prismatic VLM (prism-dinosiglip-224px+7b / Llama-2 family), so inherited Llama-2 terms must be reviewed for commercial deployment
- Safety and limitations
- Base checkpoint operates in latent action space and must be post-trained with an embodiment-specific action decoder before hardware control.
- Upstream terms of the Prismatic VLM backbone and pretraining datasets (Open X-Embodiment, Ego4D) were not excerpted in this review; impact of Ego4D dataset-access terms on checkpoint-only deployment is not established here.
Inferred from upstream model card scope, pretraining distribution, and documented deployment prerequisites.
Compute & fine-tuning
- Inference compute
- Precision:
bf16 - Edge / consumer capable: No
Exact min_vram_gb and latency_ms not stated on model card; 14.05 GB BF16 weights preclude small edge devices without quantization.
[2] · Editorial assessment
- Precision:
- Fine-tuning
- Supported: Yes
- Methods: embodiment-specific action decoder (~12M params) + LoRA (rank 32, ~123M total trainable params in upstream recipe)
- Recipe: Upstream fine-tuning documentation
- Host computer
14.05 GB BF16 checkpoint requires >=16–24 GB VRAM GPU for inference; exact minimum VRAM and multi-GPU fine-tuning specs not pinned in HF card.
Inferred from 7.54B BF16 safetensors footprint (14.05 GB).
[2] · Editorial assessment
- Operating system
- Linux
Inferred from documented Linux/CUDA Python environment and shell instructions.
[4] · Editorial assessment
Setup & usage
Open the upstream getting-started guide [4] · Model card / documentation [1] [2]
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 |
|---|---|---|
| LIBERO, CALVIN, Room2Room, SimplerEnv, real-world manipulation (arXiv:2505.06111) | See OpenDriveLab/UniVLA README & arXiv:2505.06111 | Upstream-reported simulation and real-robot evaluation claims; not independently verified by OSR. |
Project notes
Latent action space requires post-training adaptation
During pretraining, UniVLA predicts discrete embodiment-agnostic latent action tokens rather than direct robot joint commands. Deploying the ungated BF16 safetensors checkpoint on a physical arm or navigation platform follows a staged workflow documented in the pinned README's post-training section: start from the pretrained latent-action policy, attach an embodiment-specific action decoder head, and run downstream fine-tuning on target demonstrations. In upstream's reported recipe, the action decoder adds roughly 12M parameters and is trained alongside LoRA (rank 32) for approximately 123M total trainable parameters; treat these as upstream recipe figures rather than universal requirements, and do not infer minimum VRAM or training cost from them. Upstream does not pin a single minimum VRAM or closed-loop latency figure in the model card, so hardware sizing must be validated locally before closed-loop execution. [2] [4]
Upstream-reported benchmark coverage
The repository README and arXiv:2505.06111 report evaluations across LIBERO, CALVIN, Room2Room navigation, SimplerEnv, and real-world tabletop manipulation tasks. All reported results are strictly upstream-reported; Open Source Robotics has not executed training, post-training, inference, or robot trials for UniVLA and does not independently verify task performance or control rates. [4]
Separate code, weights, backbone, and dataset terms
The OpenDriveLab/UniVLA GitHub repository carries a verified root Apache-2.0 LICENSE, and the qwbu/univla-7b Hugging Face model card declares apache-2.0. Separate from those confirmed Apache-2.0 declarations, the pretrained checkpoint derives from TRI-ML/prismatic-vlms/prism-dinosiglip-224px+7b and pretraining draws on Open X-Embodiment and Ego4D. The upstream terms of that backbone and those datasets were not excerpted in this review and require independent verification; in particular, this review does not establish how Ego4D's dataset-access terms affect checkpoint-only deployment. [1] [2] [3] [4]
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
- Nov 19, 2025 [1]
- Latest release
- Aug 2, 2025 [2]
- Profile updated
- Nov 21, 2026
- Sources last checked
- Nov 21, 2026
- Next review
- May 21, 2027
- Reviewed revision
- OpenDriveLab/UniVLA@0ab9e9dd3074ffa14fdd1a00aa2aa84e4f8e1352; HF qwbu/univla-7b@968e2896809c00968e8fd936e052476ce5ac1bd0 · software repository
Separate license matrix
Open-weights 7.54B generalist VLA with Apache-2.0 codebase and ungated safetensors weights demonstrating cross-embodiment latent-action pretraining; inherited backbone and Ego4D data terms disclosed explicitly.
| Layer | Availability / license |
|---|---|
| Code | available · Apache-2.0 Apache License Version 2.0 verified in root LICENSE. [1] [3] |
| Model weights | available · Apache-2.0 (HF card declaration; upstream terms of base backbone TRI-ML/prismatic-vlms/prism-dinosiglip-224px+7b require independent verification) HF card declares license: apache-2.0; fine-tuned from TRI-ML/prismatic-vlms/prism-dinosiglip-224px+7b, whose upstream terms were not excerpted in this review. [2] [4] |
| Training datasets | available · Mixed upstream (Open X-Embodiment + Ego4D; upstream dataset terms not excerpted in this review) [2] [4] |
| Documentation | available · Apache-2.0 [3] [4] |
References
Each source links to the exact revision we read. “Used for” shows where this page relies on it.
- S001 OpenDriveLab/UniVLA GitHub metadata and HEAD commit
Default branch main HEAD is 0ab9e9dd3074ffa14fdd1a00aa2aa84e4f8e1352 (2025-11-19T03:31:23Z); SPDX license Apache-2.0.Revision, date & where cited
OpenDriveLab (GitHub) · github.com · revision
0ab9e9d· GitHub API repos/OpenDriveLab/UniVLA & commits/main · Accessed Nov 21, 2026
Used for: Overview, Setup & usage, Separate code, weights, backbone, and dataset terms, What we checked and 1 more - S002 Hugging Face model card and API metadata: qwbu/univla-7b
7,541,237,184 BF16 parameters, 14.05 GB safetensors, ungated, license: apache-2.0, lastModified 2025-08-02; pretrained on OpenX + Ego4D.Revision, date & where cited
qwbu / OpenDriveLab (Hugging Face) · huggingface.co · HF API & README.md at revision 968e2896809c · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Checkpoints, Is this for you? and 6 more - S003 OpenDriveLab/UniVLA LICENSE at pinned commit
Revision, date & where cited
github.com · revision
0ab9e9d· LICENSE at commit 0ab9e9dd3074 · Accessed Nov 21, 2026
Used for: Separate code, weights, backbone, and dataset terms, Separate license matrix - S004 OpenDriveLab/UniVLA README.md at pinned commit
Documents Model Zoo (LAM stage 1/2, univla-7b with Prismatic VLM backbone), LIBERO, CALVIN, Room2Room, SimplerEnv, and real-world experiments.Revision, date & where cited
OpenDriveLab (GitHub) · github.com · revision
0ab9e9d· README.md at commit 0ab9e9dd3074 · Accessed Nov 21, 2026
Used for: Overview, Model card summary, Checkpoints, Is this for you? and 7 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 draft prepared from pinned GitHub commit 0ab9e9dd3074 and HF revision 968e2896809c.