GeneralSequences

Products

Inference Physical AI models on GPUs close to your robots: start from a template like π0.5, or bring a policy you trained. Benchmark Benchmark a policy automatically on simulations of its target environment, starting from a template suite.

Customized models

Start from a template like π0.5 or bring a policy you trained — it runs on your own deployment.

Browse templates →

Models

In the library

Physical AI templates π0.5, GR00T and Cosmos policies to deploy as your own, on GPUs close to your robots. Vision-Language Model VLM Claude, GPT, Gemini, Qwen, Perceptron, GLM — some of them self-hostable. Decision DEC A state and typed questions in, calibrated probabilities out. Not chat — for routing, ranking and verification. Benchmark templates LIBERO, RoboCasa and RoboTwin — the simulations a policy is scored on.

Model library

Every model and template: physical AI templates you deploy, the vision-language and decision models a policy can call, and the benchmark suites it is scored on.

View all models →

Developers Pricing

Company

About Who we are and what we build. Security How we protect your workloads and data, and how to report a vulnerability. Status Whether the API, the console and the workers are up, checked every minute. Contact us Send us your name, email and a message. Legal Terms of Service, Privacy Policy and Acceptable Use Policy.

Products

Inference Benchmark

Models

All models and templates Physical AI templates Vision-Language Model Decision Benchmark templates

Developers Pricing

Company

About Security Status Contact us Legal

Status

Whether each part of Sequences is working, checked every minute at the addresses you use. GPU workers show the status our compute provider reports.

Loading the current status…

General Sequences · 2026 Sign in About Security Status Legal