Alternatives to Mrpt

Open-source research framework that adapts text-to-image diffusion models to follow natural-language instructions across vision tasks.

Physical Intelligence builds vision-language-action robot foundation models that control any robot for any manipulation task.

Open-source autonomous driving simulator with a Python API, ROS2 support, and Unreal Engine 5.5 rendering.

Autodistill auto-labels image datasets with foundation models, then trains small vision models you own.

Open-source procedural generation that renders photorealistic 3D worlds and labeled synthetic training data from Python code alone.

Skyfall-GS turns multi-view satellite imagery into free-flight navigable 3D city blocks with 3D Gaussian Splatting and diffusion refinement.

Open-core LiDAR SLAM for ROS 2: LiDAR odometry, LIO, 3D SLAM and georeferencing you can modify at the source level.

OpenCOOD is an open cooperative perception benchmark for V2V 3D detection and LiDAR fusion research.

Open-source Unreal Engine plugin that turns UE4/UE5 scenes into labeled synthetic image datasets for computer vision.

Open-source Python SDK and CLI for training and deploying physical AI models — VLA and SLM — from managed cloud compute to real robot hardware.

Skild AI is building an omni-bodied robotic brain to control any robot for any physical task.

Per-plant crop intelligence platform from Alphabet's X, now acquired by Driscoll's and John Deere

Rhoda AI builds video-predictive robot foundation models that learn heavy industrial manipulation from a single human demo.

AI facial analysis scoring 160+ beauty markers into a personalized, non-surgical glow-up protocol backed by named research.

License the Nuro Driver for L4 autonomous driving across robotaxis, commercial fleets, and personal vehicles.

Wayve AI Driver: mapless, vehicle-agnostic autonomy that learns to drive from data, from L2 to L4.

CVAT is an open-source data annotation platform for image, video, 3D point cloud, and audio labeling.

Offline AI image annotation with local SAM2, SAM3 and YOLO-World — one-time purchase from $49.

Build, code, and program OpenCat quadruped robot kits — Bittle, Nybble Q, and the new Quaddle desk robot dog.

Cerul is semantic video search: ask in plain words, get the exact timestamped moment with citable evidence.

Janus Pro unifies multimodal image understanding and text-to-image generation in one open-source DeepSeek model.

Meta's open-source promptable segmentation model for selecting and tracking objects in images and video

Open-source simulated benchmark for language-conditioned, long-horizon robot manipulation from the University of Freiburg.

Open-source Java software for processing and analyzing scientific images, extensible through hundreds of plugins

SAM 3 segments every matching object in images and video from a text or visual prompt — free to try in the Segment Anything Playground.

Odyssey-2 Max is a causal world model that simulates physical scenes in real time as you feed it actions.

MagicDrive is an open-source framework that generates multi-camera street-view images and video from precise 3D geometry control for autonomous driving

6thSense captures tactile sensing and egocentric touch data for training dexterous robot policies

Cortex AI supplies real-workplace egocentric video and robot trajectory data for training embodied AI and robot foundation models.

Free browser tool that identifies Chinese, Japanese, and Korean typefaces from an uploaded image and returns confidence scores plus weight, slant, and stroke