Documentation
EsaSim documentation
Start with the first three pages: they take you from a new account to a simulation running on your own computer. Everything else is here when you need it.
01Getting startedAccount, GitHub, your computer and a first run, in about fifteen minutes.
02How EsaSim worksProjects, blueprints, blocks, workers and runs, and where everything lives.
03Your first studyAsk a question, and an AI assistant builds and runs the simulation that answers it.
Looking for a block? The block reference lists every one, with its inputs, outputs and settings.
Get started
Getting startedFrom a new account to a first run on your own computer, in four steps and about fifteen minutes.
How EsaSim worksThe few ideas you need: projects and blueprints, the workers that run them, and where everything lives.
Set up a workerInstall the EsaSim worker on any computer you run, here or remote, and set it up from the dashboard.
Tutorials
Tutorial: build a study with an AI assistantFrom a question to an answer in one conversation: connect an assistant, let it build and run a drone landing study on your worker, read the results and improve the study.
Tutorial: a project in codeBuild a project as files in your own editor: a worker block in Python, a blueprint in JSON, pushed to GitHub, imported, run, and improved with git.
Tutorial: from a blueprint to your appBuild a blueprint that reviews pictures, test it on cases you know, publish it as an endpoint and call it from your own app.
Tutorial: blocks that work togetherWrite a worker block in Python and a view in JavaScript, and connect them: a CPU benchmark shown as text, live values while it runs, a Stop button that talks back to it, and blocks on two workers feeding one comparison.
Demo projects
Demo projectsComplete projects to get started: what each demo shows, its flows, what it needs and how to run it on your workers.
Text to imageOpen text-to-image models turn one prompt into any number of pictures on your worker, shown live as they are made, and compared from small to large.
Text to videoOpen text-to-video models make short clips on your NVIDIA GPU, each streamed live and played whole as it is made; keep the good ones and browse them.
Data curationFrom a raw pile of images to a training-ready set: quality, duplicates, embeddings, label checks, a review grid, a split and a before and after report.
CT to 3D anatomyCT scans become coloured 3D models of every organ, bone and vessel to turn around, cut through and print, and a soft-tissue model that moves under gravity.
Warehouse fleetA goods-to-person warehouse run live in 3D with up to fifty robots, then sweeps that find how many robots it needs and which layout works best.
Booster landingA rocket booster learns to land on a drone ship by reinforcement learning on your GPU, then launches and lands live in 3D with onboard cameras.
Build
ProjectsProject spaces: settings, data, team and roles, blueprints, the workers a project may use, runs and results, and analytics.
BlueprintsBuild a flow from blocks on a grid, run it on your workers, and watch live telemetry and results on the View page, in charts, tables and the 3D view.
AI assistants (MCP)Connect Claude, ChatGPT, Cursor or any MCP client to your workspace so it can read your projects, build flows and blocks, run them on your workers and read the results.
Build your own blocksMake a block of your own in Python (runs on your workers) or JavaScript (a view in the browser): its ports and their formats, its settings, its code, versions, sharing with the EsaSim library and copies on GitHub.
Data and datasetsConnect where your data lives (folders on your workers, S3 buckets, Google Drive, Dropbox, OneDrive), upload and download files, browse them with a viewer for each kind, and send results from flows to datasets as they are made.
The EsaSim libraryFree test data for every workspace: browse 3D models, pictures, videos, logs and tables, and use them in flows with the Library block. For platform admins: adding collections.
Guides
AI agents and modelsThe LLM and Agent blocks: connect a model (EsaSim, a provider with your key, or your own server) and let an agent review images, logs and tables in a loop.
Images and videos from textGenerate pictures and short videos on your own workers with open models from Hugging Face, watch them arrive live, keep the good ones and save them as a dataset.
3D scenes and renderingSay which model each object uses and which trajectory it follows, replay the scene in the 3D view, and render it on a worker into pictures and videos.
Robots: from simulation to real hardwareTraining robots in simulation with EsaSim, what makes a policy work on the real robot, and where EsaSim is going: robot descriptions, robot categories and bridge workers that run policies on real devices.
Run and ship
Runs and jobsRunning a blueprint: the run panel, placing blocks on several workers, saved profiles, tags, where a run is, data copied to the worker that needs it, the time limit, the queue, and the sessions coming next.
Results and usageWhere job outputs are stored, how long results are kept, and how to read the usage page.
API endpointsPublish a blueprint as an endpoint so your own app, service or CI can run it: send inputs, read the results as JSON.
Workers
Shared GPU serversInstall the EsaSim agent on a Linux server with NVIDIA, AMD, Intel, Huawei Ascend or Tenstorrent accelerators and Docker.
Compute unitsSplit a multi-GPU server into isolated slices of GPUs, RAM, disk and CPU cores for different workloads.
Teams and account
Organizations and rolesCreate an organization for your company, lab or team, choose its plan, invite members, assign roles and teams, and set per-member limits.
Security settingsTwo-factor authentication, session timeout, allowed IP ranges, audit log and signed-in devices.
Plans and billingCompare plans, upgrade, manage seats and billing interval, and handle invoices and payments.
Cloud creditsIn development: cloud GPUs paid with prepaid credits, and how reservations, per-second charges and refunds of unused time will work.