Laboratory
Entity that develops or publishes the model and maintains its technical and safety documentation.
Explore the documented models, published evaluations and primary sources associated with this organisation.
Runway video-and-action policy adapted to LeRobot’s SO-100 / SO-101 follower arm. The checkpoint appeared on November 12 and the official code added its configuration that day. It controls joints and the gripper; it is a distinct adaptation from DROID and YAM.
13 NOV 2026↗ BASE FOR ADAPTING ROBOT POLICIESBase checkpoint of Praxis-1, Runway’s video-action model family made public on November 11. It includes the MoE diffusion trunk and action expert but lacks heads for any particular robot.
12 NOV 2026↗ DROID SINGLE-ARM CONTROLPraxis-1 policy fine-tuned for a Franka arm with a Robotiq gripper and exterior and wrist cameras. It predicts video and actions jointly in 1.6-second chunks. BF16 weights are available under Apache 2.0.
12 NOV 2026↗ YAM BIMANUAL CONTROLPraxis-1 policy fine-tuned for two YAM arms with a top camera and two wrist cameras. It produces video and action chunks for both arms, using poses relative to each chunk’s start and control at 15 Hz.
12 NOV 2026↗Figures are shown with the context reported by their source. A provider result is not an independent comparison and does not replace your own evaluation.
| Model | Benchmark | Result | Metric |
|---|---|---|---|
| Praxis-1 DROID | RoboLab-120 | 46,2 % | Simulation success |
An organization, a product, and a model are not the same unit. Inferama separates them to avoid attributing capabilities or commercial terms to the wrong item.
Entity that develops or publishes the model and maintains its technical and safety documentation.
Identifiable version with limits, modalities, and behavior that may change between releases.
API, application, associated cloud, or commercial plan; each channel may have different pricing, retention, and limits.
Every claim must retain the official page consulted and the verification date.
Praxis-1 YAM combines video prediction and actions to control two YAM arms. The published checkpoint requires an 80 GB GPU in its reference configuration; the NVFP4 option reduces memory on certain GPUs but is experimental and does not equate to a real-world control evaluation.
15 Nov 2026 ↗ NOTICIADated November 12, 2026, the checkpoint controls SO-100/SO-101 arms through joint-space actions. The reference version occupies 54.4 GB and calls for an 80 GB GPU; its NVFP4 variant reduces inference memory experimentally but downloads the same complete weights.
13 Nov 2026 ↗