Generalist AI has released GEN-1.5, a robot foundation model that learns a new physical task from a single demonstration [reference:14]. Drop 3-12 seconds of sensorimotor data into its 30-second context window, and the robot performs the task. No gradient updates, no fine-tuning, no task-specific programming [reference:15].
One-shot in-context learning
Across 10 diverse manipulation tasks, one-shot in-context prompting averaged 59% success (±10% std. dev.) straight from the pretrained model. Ten gradient steps on five minutes of data per task raised that to 83% (±9%). Generalist calls the mechanism "in-context sensorimotor learning," and says it was never trained for — no architectural changes, no auxiliary objectives. It emerged from over eight months of continuous pretraining on physical interaction data [reference:16].
Technical details
GEN-1.5 is a large multimodal model that takes video, sensor, language, and proprioceptive inputs, holds 30 seconds of memory, and emits 100 Hz action trajectories [reference:17]. It has been pretraining continuously for over eight months on physical interaction data captured in homes, warehouses, and factories.
The main mechanism is in-context sensorimotor learning. A sensorimotor example — sensor streams plus the action trajectory — is inserted into the 30-second context window through a drag-and-drop interface. The model then performs the task immediately, with zero gradient steps and no fine-tuning.
Emergent capabilities
Crucially, none of this was designed in. Generalist states there were no architectural changes to promote in-context learning, no special training objectives, and no auxiliary objectives encouraging improvisation. The capability emerged from pretraining scale, the same way one-shot prompting emerged in large language models.
Light fine-tuning results
Adapting robot policies has typically taken tens of thousands of gradient steps. Ten steps here move the model weights on held-out tasks by less than 0.15%, which suggests fine-tuning is reconfiguring knowledge the model already has rather than building new representations. Generalist frames it as "adaptation in an extremely low-data regime."
Generalization examples
Two independently recorded prompts placed in context get chained into one continuous behavior. The model produces the bridging motions — repositioning, regrasping, error recovery — that appear in neither demonstration. A demonstration recorded entirely in simulation works as a prompt for the real robot, despite pretraining containing no simulation data. In some cases a person demonstrates with their own hands, in view of the robot's cameras, and the model reproduces it with the robot's hands.
After light fine-tuning, trained on five minutes of brushing a block into a bowl, the model used a banana as a makeshift brush, and used a dustpan to lift and dump the block instead — a different contact sequence entirely. It also removed a sheet of paper covering the bowl, and worked ambidextrously when demonstrations used one hand.
Availability
There are no public weights, no API, no pricing page, and no self-serve product. Generalist AI runs GEN-1.5 on its own fleet and data engine. Anyone who wants it today goes through a direct partnership.