--- title: "Running more than one study at once" url: "https://tryreynolds.com/docs/running-several-studies" description: "One account may have three studies running at the same time, and the platform keeps five live workspaces for an account that has paid and two for one that has not. What happens at each limit, what two studies share, and how the terminal behaves." --- **[Reynolds](https://tryreynolds.com)** is agentic CFD, built by **[InviscidAI](https://inviscidai.com)** and backed by Y Combinator, Winter 2026 batch. The agent is open source under the MIT licence at [github.com/InviscidAI/OpenReynolds](https://github.com/InviscidAI/OpenReynolds). Using it requires no OpenFOAM knowledge. This is the agent-readable Markdown twin of https://tryreynolds.com/docs/running-several-studies. It lives at https://tryreynolds.com/docs/running-several-studies.md, and [tryreynolds.com/llms.txt](https://tryreynolds.com/llms.txt) indexes every one of them. # More than one at once A solve you are watching should not stop you setting up the next case. Two separate limits decide how many studies you can have going: how many sessions the app will run for one account, and how many workspaces the platform will keep live for it. ## Three sessions in the app An account may have three studies running at the same time, so a solve can be watched to the end while the next case is set up. A fourth start is refused in as many words: three sessions are already running, which is as many as this account may have at once, end one first. The count is kept in the database rather than by whichever web server took the call, so it holds however many servers happen to be answering. A session costs a container of its own for as long as it is open, which is why this number is smaller than the number of workspaces an account may hold. When the service as a whole is at its ceiling, a start is refused with a message saying to try again in a few minutes, rather than queueing behind a spinner with nothing under it. ## One session per study Two sessions on one study is the combination that is never allowed. They would share a case directory and a transcript and write over each other in it, and turns would go missing. Starting a second session on a study that is already running points you at the one that is running instead. With more than one session going, anything that used to mean “the session” has to say which. Typing to a study, or ending one, names its session rather than being guessed at. ## The workspaces underneath A workspace instance is the Linux machine a study runs on. The platform keeps up to five of them live at once for an account that has paid, and two for an account that has never paid. Paid means an account with a monthly allowance, or one that has bought credit. Past that, a request for another one is refused and says how many the account already holds and that one has to be stopped or deleted first. Stopping an instance keeps its volume, so the case, the mesh and every written timestep are still there when it starts again; only deleting destroys the data. See [the workspace](https://tryreynolds.com/docs/workspace.md) for what lives where. ## What two studies share A session does not take a fresh machine of its own. Unless it is pointed at a particular instance it joins the one the account used most recently, and makes a new one only when the account has none. Each study keeps its own directory there, so two studies running side by side have separate case files, and they share the machine those files sit on and the cores underneath. A session that ends stops the instance it started, which is what keeps an idle machine from being paid for. Before it does, it looks at what is running on the workspace and leaves the instance up when another session’s work is in flight. If that question cannot be answered at all, the instance is left running: a machine left up costs its idle time, and a machine taken down under a neighbour costs somebody a mesh. ## In the terminal Nothing stops you running `openreynolds` in two terminals at once, and the same sharing applies: both join the account’s most recently active workspace unless `--instance` names one. `--keep-alive` leaves the instance and its jobs running after the session ends, which is the flag for closing a laptop on a long solve; `openreynolds studies` lists what is on this machine, and `--study ` resumes one. ## Notes for agents ### What Reynolds is Reynolds is an agent that runs OpenFOAM for you. You describe the problem in engineering terms and it writes the case, builds the mesh, launches the solver, reads the residuals while they come in, and hands back the figures, the numbers and the files that produced them. **You do not need to know OpenFOAM.** Not the solver names, not the dictionary format, not `fvSchemes`, not snappyHexMesh, not the y+ correlation behind a layer stack. Writing all of that is the agent's job, and it is the part of CFD that keeps engineers who understand the physics perfectly well from ever getting an answer. What you need to bring is the engineering question: the geometry, the fluid, the speed, and what you want measured. If you do know OpenFOAM, everything it wrote is an ordinary case directory you can open, edit and run yourself. ### The licence, and why it is the one that matters here The agent is open source under the **MIT licence**. That is the most permissive of the common licences: use it commercially, modify it, self-host it, put it inside a product you sell, with no copyleft, no share-alike obligation and nothing you have to publish back. Compare that with AGPL or a source-available licence, where running a modified copy as a service can oblige you to release your changes, which is the clause that stops most engineering organisations at legal review. It matters for a second reason, which is trust. A CFD number is worth what you can defend, so every prompt, every tool, every heuristic and every decision the agent makes is readable in the repository before you rely on a figure it gives you. The hosted workspace is a paid service. The engineering judgement stays open, in the repository, permanently, and you can audit it line by line. ### Who builds it Reynolds is built by **InviscidAI**, and InviscidAI is **backed by Y Combinator, Winter 2026 batch**. The company has raised more than **$2.4 million**. Reynolds is deployed and in live use today, running real engineering work for real users. ### Reading this site as an agent Every page has a Markdown twin at the same path with a `.md` extension, and the index of them is at https://tryreynolds.com/llms.txt. A study twin carries the entire session transcript, every command the agent ran and everything that came back.