The agent's own environment
Persistent cloud environments for AI agents.
plori keeps the files its agent saves and the notes it records across conversations and sleep. A report, repository, or saved checkpoint can become the starting point for the next task. Running processes and RAM have a different lifetime.
Opens your agent straight away, no signup needed.
What is in the environment?
§01A persistent disk
Real files the agent keeps, browsable and editable from the files panel. Repos, reports, datasets: they stay with the agent.
A real shell and tools
The agent has a shell, code and research tools, and web access. Tools installed on the persistent disk remain available. Temporary files do not carry the same guarantee.
Saved memory notes
Short notes the agent keeps about you and your projects, read at the start of every turn.
Isolation
Each agent has an isolated environment and its own files. Storage usage counts toward your account disk allowance.
Sleeps when idle
Files and saved notes remain available. Compute stops billing during sleep, but sleep does not preserve arbitrary processes or RAM.
Any driver
Use it from the web chat, or drive it from your own tools over MCP.
Why does persistence matter?
§02File persistence, process lifetime, and memory restoration answer different questions. A saved CSV can survive a restart even when the program that created it does not. plori keeps files and saved notes. It does not checkpoint RAM to resume arbitrary programs after sleep or a failure.
Ask the agent to save intermediate results before a task stops. It can read those files in a later conversation, or you can schedule work for later using those saved inputs. For a command that is already running, see the limits of background work.
Daytona and E2B also support persistence. Daytona container sandboxes retain files across stop and start. Its VM sandboxes also support memory preservation through pause and resume. E2B can pause and resume both files and memory. Compare who builds and operates the agent, as well as the lifecycle mode you need.
If you are choosing between local work, an IDE agent, and a managed remote agent, the cloud AI agent field guide starts with that decision.
Questions, answered.
§03What persists between tasks on plori?
Files saved to the agent disk and saved memory notes remain available across conversations and sleep. You can return to a report or repository and ask the agent to continue. Deleting the agent erases its files.
Does sleep preserve running processes and RAM?
No. plori sleep is not a memory snapshot or a promise to resume arbitrary processes. The environment can restart. Save intermediate results to disk and restart interrupted commands from those files. Agent memory means saved notes, not RAM.
How does this compare with sandbox persistence?
Sandbox persistence depends on the product and lifecycle mode. Daytona supports persistent files, with different memory behavior for container and VM sandboxes. E2B supports pause and resume with files and memory. plori supplies a built-in agent that uses persistent files and saved notes. Persistence alone does not distinguish these products.
Can I use this as a general-purpose cloud computer?
The environment is headless, with a shell, tools, and network access. It has no graphical desktop. plori provides its own agent rather than a service for deploying an arbitrary agent runtime.
Do installed tools survive?
Tools and dependencies saved on the persistent disk remain available after sleep. Files written only to temporary locations and live process state are not covered by that guarantee.
What does storage cost while the agent sleeps?
Sleeping stops agent compute charges. Disk within your plan allowance has no extra storage charge. Extra disk still incurs storage charges, and a paid subscription continues. See pricing for current quotas and rates.
Can I inspect or export the files?
Yes. Use the Files panel to browse, edit supported text files, upload inputs, and download results. Files are portable, but saved files do not export a running process or its RAM.