Vision-language-action model (VLA)
Also known as: VLA, VLA model, vision-language-action
A vision-language-action model (VLA) is an AI model that combines what a robot sees (vision), instructions in plain language, and the movements needed to act (actions) in one system.
You might tell a robot 'put the apple in the bowl,' and the VLA turns camera input and that sentence into arm and gripper commands. While VLAs have made robots far better at handling new objects and instructions, they still fail in unpredictable ways and need huge amounts of robot training data.
Vision-language-action model (VLA), explained at three levels
At home
A VLA lets you talk to a robot in plain words. The robot looks through its cameras, understands what you asked, and figures out how to move its arm to do it.
At work
VLAs could let businesses retrain robots for new tasks by giving instructions and examples, rather than hiring programmers to write new code.
At home
VLAs build on the same kind of AI as chatbots, but instead of only producing words, they produce movements. They learn from many recordings of robots, and sometimes people, doing tasks.
At work
A VLA typically starts from a pretrained vision-language model, then is fine-tuned on robot demonstration data to output action tokens or continuous motor commands. Some systems pair a slower 'reasoning' model with a fast low-level controller.
At home
These models can handle objects they've never seen before, which older robots couldn't. In my opinion, the catch is they can also fail on simple things in surprising ways, so they need careful testing.
At work
Challenges include scarce robot data compared with internet text, latency for real-time control, safety guarantees, evaluation across environments, and transfer between robot bodies. Teleoperation, simulation, and human video are all used to scale training data.
Real-world examples
At home
Aisha visited a university robotics lab where a researcher typed 'hand me the blue cup,' and a robot arm picked the right cup from a cluttered table, despite never seeing that cup before.
At work
Thomas runs automation at a Quebec food packaging plant. He's testing a VLA-powered robot arm that can be taught new packing patterns from a few demonstrations instead of reprogramming.
How it compares
| feature | Vision-language-action model (VLA) | Traditional robot program |
|---|---|---|
| Input | Images and language | Coded instructions and positions |
| New objects | Can generalize | Must be programmed |
| Predictability | Lower | High |
| Setup for new tasks | Examples or instructions | Engineering work |
VLAs trade predictability for flexibility.
For analysts
VLA-based robots can cut reprogramming costs for varied tasks. I'd keep them in fenced or supervised settings until their failure rate on your actual task is measured.
History & origin
Earlier robots relied on hand-written programs for each task. Researchers began training robots on large datasets in the 2010s. Google's RT-1 in 2022 trained a transformer on robot demonstrations, and Google DeepMind's RT-2, announced in July 2023, coined the term vision-language-action model by building on a vision-language model trained on web data. The Open X-Embodiment dataset, released in 2023 by dozens of labs, pooled data from many robot types. OpenVLA followed in 2024 as an open model, Physical Intelligence released its pi-zero (π0) model in late 2024, and in 2025 Figure showed Helix, NVIDIA released GR00T N1 for humanoids, and Google DeepMind launched Gemini Robotics. In my view, VLAs are the first robotics approach that makes the 'general-purpose robot' idea seem plausible, though still years from routine.