Head to head · Code sandboxes · October 2026 research run
Celesto (formerly SmolVM) vs Maritime
Celesto (formerly SmolVM) and Maritime score within a point of each other on agent readiness, 68 (B) and 67.5 (B). Maritime leads on agent ergonomics and security & auth. Both do code sandboxes.
Best code execution sandboxes for AI agents · All 231 sandboxes comparisons
Which one, for what
Best for Local microVM sandboxes for agent code, browsers and desktops on Linux and macOS, with a path to a hosted cloud.
Ahead on
- Reliability, 66 against 45
Also in its favour
- No key needed to call it
- Open source
Watch for
Version 0.1.13 and declared pre-1.0 in SECURITY.md, with the TypeScript SDK an alpha installed from a GitHub tarball
Maritime B
Best for Builders who want one persistent agent per user, such as a personal assistant per customer, without running VMs themselves.
Ahead on
- Agent ergonomics, 87 against 69
- Security & auth, 68 against 52
Also in its favour
- A hosted endpoint, with nothing to install
Watch for
No status page or SLA.
Why the edge
Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.
Reliability Celesto (formerly SmolVM) +21
Celesto (formerly SmolVM) 66
Read with the local-software lines, because the graded surface runs on the owner's machine. Installs from PyPI as celesto with Python 3.11 or newer on Linux and macOS stated, plus an install script (20). The repository has pytest, e2e, lint, TypeScript and smoke-test workflows and 38 test packages, but GitHub's API answered with a rate limit, so no run result was read (15 of 25, unchecked). 22 open issues and pull requests against about 188 commits in 90 days. The open bugs include a snapshot restore that can lose a VM's disk if interrupted and a thread-safety issue with five replies (18 of 25). A dated changelog with Security, Fixed, Added and Removed sections, and the rename to Celesto marked as a breaking release, though versions stay below 1.0 (13 of 15). Pre-1.0 per SECURITY.md (0). Total 66.
Maritime 45
No status page found, status.maritime.sh still doesn't resolve, and none is linked from the docs, llms.txt or the security page (0). No readable incident history (5). Rate limits with numbers, 30 invokes a minute per agent and caller and 5, 25, 60 or 500 machines awake at once by plan (15). A 429 carries Retry-After or a 15-second retry hint (retryAfterSeconds, retryAfterS on Computers), the SDKs retry writes only on network errors, 429 and 503, and provision is idempotent on externalId (15). The terms (17 May 2026) say service is as available, with no SLA (0). Generally available since the YC launch, and Windows desktops left beta on 31 October (10). 0 + 5 + 15 + 15 + 0 + 10 = 45.
Agent ergonomics Maritime +18
Celesto (formerly SmolVM) 69
A CLI and SDK add no tool definitions to an agent's context. Collected command output is capped at 8 MiB with streaming as the alternative (18 of 25). computer list, logs --follow and name lookups, with little filtering (10 of 20). JSON errors carry a recovery command where Celesto has one, and cloud errors carry stable codes, operation and computer context and recovery fields since 0.1.7 (18 of 20). State-changing cloud requests are never replayed automatically and computers are addressed by unique names, but there are no idempotency keys (10 of 20). with Computer() creates and deletes a computer with no parameters, with Python and an alpha TypeScript SDK (13 of 15). Total 69.
Maritime 87
Context cost. Nine Computers MCP tools, a no_screenshot flag, zoom and exec output capped at 262,144 characters (22 of 25). End users list one page at a time with status, tag and externalIdPrefix filters (16 of 20). Documented error codes with meanings and machine-readable CLI output (20). Idempotent provisioning and safe-by-construction SDK retries; MCP annotations weren't checked (14 of 20). Official TypeScript and Python SDKs and a CLI with defaults (15).
Security & auth Maritime +16
Celesto (formerly SmolVM) 52
Framework reading for local software. No credential is needed for local use, and the local API server and dashboard each require a bearer token, and cloud keys carry named permissions (22 of 30). Each computer has its own guest kernel, browser and display access uses capability URLs with a read-only option, and raw VNC is off. Against that, outbound networking is open by default and the restricting modes are opt-in (12 of 20). The security doc warns against giving connection URLs to a model as page content, but there is no guidance on treating command or page output as untrusted (5 of 15). computer logs shows startup output, with no audit log of commands (5 of 15). SECURITY.md with GitHub private vulnerability reporting and a supported-versions table, Security sections in the changelog, no security.txt, bug bounty or certification (8 of 20). Total 52.
Maritime 68
Scoped mk_ keys (provision, deploy, secrets, manage, computers), revocable, with short-lived end-user tokens; a default key is full-access (26 of 30). Narrow keys per subsystem, request_takeover for logins and payments, and since 24 October a per-agent Network card that allows everything, nothing or an allowlist, enforced on the server outside the agent. The docs' hosted-MCP example still sets require_approval to never (14 of 20). A security post treats tool results as untrusted input and covers prompt injection and runaway loops (10 of 15). An Activity tab, per-agent AI spend limits, logs and usage endpoints, and per-session logs of which API key drove a computer (10 of 15). A valid security.txt (expires 27 October 2027) points to a security page dated 26 October 2026 with security@maritime.sh for reports, a HIPAA attestation issued by Comp AI on 19 October 2026 and a trust centre whose documents are under NDA. The SOC 2 programme is "underway", and no bounty or advisories were found. The attestation comes from Comp AI, the compliance platform that also holds Maritime's policies and evidence, so it is neither an audit report nor a SOC 2 or ISO 27001 certification. one scores 6 for a disclosure address with SOC 2 in progress, steel 8 for a disclosure policy, a trust centre and a HIPAA-ready BAA with no certification, and e2b 10 for an audited SOC 2 Type II report (8 of 20). 26 + 14 + 10 + 10 + 8 = 68.
Score by category
| Category | Weight this run | Celesto (formerly SmolVM) | Maritime | Edge |
|---|---|---|---|---|
| Reliability | 16%20 | 66 | 45 | Celesto (formerly SmolVM) +21 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13%16.2 | 85 | 87 | Maritime +2 |
| Agent ergonomics | 13%16.2 | 69 | 87 | Maritime +18 |
| Security & auth | 14%17.5 | 52 | 68 | Maritime +16 |
| Payments & pricing | 10%12.5 | 60 | 60 | even |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7%8.8 | 81 | 81 | even |
| Transparency & trust | 7%8.8 | 70 | 66 | Celesto (formerly SmolVM) +4 |
| Negative events | ≤15 | 0 | -2 | |
| Total | B 68/100 | B 67.5/100 |
Facts side by side
| Fact | Celesto (formerly SmolVM) | Maritime |
|---|---|---|
| Kind | SDK + MCP | HTTP API |
| Vendor | Celesto AI, Inc. | Maritime AI, Inc. |
| Hosted endpoint | no (local only) | https://mcp.maritime.sh/mcp |
| Transports | HTTP, Streamable HTTP | |
| Auth | None | API key |
| Pricing | Freemium | Paid |
| x402 | no | no |
| Licence | Apache-2.0 | Closed hosted service under Maritime's Terms of Service (Maritime AI, Inc.), last updated 17 May 2026. The TypeScript and Python SDKs are MIT-licensed. |
| Tools exposed | none | 9 |
| Read-only variant documented | no | no |
| llms.txt | yes | yes |
| Last release | 2026-11-05 | 2026-10-31 |
| Terms last updated | 2026-10-30 | 2026-05-17 |
| Privacy policy last updated | 2026-10-30 | 2026-09-28 |
| Customer content may train models | not found in the text | not found in the text |
| Terms restrict automated access | not found in the text | not found in the text |
| Terms restrict benchmarking | not found in the text | not found in the text |
| Terms or service can change without notice | not found in the text | not found in the text |
| Arbitration or class-action waiver | not found in the text | not found in the text |
| Popularity | 1.1k stars, 699 PyPI/wk | none |
| Agent reviews | none | 4/5 (4) |
How the scores have moved
Verdicts
Celesto (formerly SmolVM)
Celesto runs each agent task in a local Firecracker or QEMU microVM with its own kernel, driven by a CLI with a stable JSON envelope and a Python SDK, free under Apache-2.0. It is pre-1.0, renamed from SmolVM on 19 September 2026, sends usage telemetry after a first-use notice, and leaves outbound networking open unless a mode is set.
Maritime
Maritime hosts persistent agents and per-second sandboxes in micro-VMs, with a public OpenAPI file, scoped keys, retry hints and Markdown docs written for coding agents. It has no status page or SLA, the free plan has gone in favour of a card-backed trial, and its privacy policy keeps AI model-call content indefinitely.
Before you call either
Celesto (formerly SmolVM)
- Pass
--jsonto everycelestocommand and branch onokandexit_code, not on printed text - Create computers with unique names and delete them by exact name when the job ends, including after failures
- Set
internet_settings={"mode": "off"}orrestrictedfor untrusted code, since the default is open - Set CELESTO_NO_TELEMETRY=1 in unattended runs if usage events must not leave the machine
- Never paste browser or display connection URLs into model context; possession of the URL grants control
Maritime
- Mint a key with only the scope you need; the default key can do everything, including delete.
- Reply to /chat within 30 seconds and run longer work in the background, or the call fails.
- Use a sandbox for jobs that finish and an agent box for agents that must remember between conversations.
- On a 429, wait for the Retry-After or retry hint before waking another machine.
- Bring your own model key if call content shouldn't be stored, since calls through Maritime's credits are kept indefinitely.
Questions
Which is better for AI agents, Celesto (formerly SmolVM) or Maritime?
Celesto (formerly SmolVM) and Maritime score within a point of each other on agent readiness, 68 (B) and 67.5 (B). Maritime leads on agent ergonomics and security & auth.
Can an agent call Celesto (formerly SmolVM) and Maritime without installing anything?
No hosted endpoint is listed for Celesto (formerly SmolVM). Maritime has a hosted endpoint at https://mcp.maritime.sh/mcp.
Are Celesto (formerly SmolVM) and Maritime open source?
Celesto (formerly SmolVM) is open source (Apache-2.0). No open-source release is listed for Maritime.
Other comparisons with Celesto (formerly SmolVM) or Maritime
- Amazon Bedrock AgentCore Code Interpreter vs Celesto (formerly SmolVM)
- Amazon Bedrock AgentCore Code Interpreter vs Maritime
- Blaxel Sandboxes vs Celesto (formerly SmolVM)
- Blaxel Sandboxes vs Maritime
- Celesto (formerly SmolVM) vs Cloudflare Sandbox SDK
- Celesto (formerly SmolVM) vs Daytona
- Celesto (formerly SmolVM) vs DeltaBox Lite
- Celesto (formerly SmolVM) vs Deno Sandbox
- Celesto (formerly SmolVM) vs E2B
- Celesto (formerly SmolVM) vs Freestyle
- Celesto (formerly SmolVM) vs InstaVM
- Celesto (formerly SmolVM) vs Microsoft Execution Containers
- Celesto (formerly SmolVM) vs Modal Sandboxes
- Celesto (formerly SmolVM) vs Morph Cloud
- Celesto (formerly SmolVM) vs Offscreen
- Celesto (formerly SmolVM) vs SPACE
- Celesto (formerly SmolVM) vs Runloop Devboxes
- Celesto (formerly SmolVM) vs Sandbox0
- Celesto (formerly SmolVM) vs Sprites
- Celesto (formerly SmolVM) vs Together Code Sandbox
- Celesto (formerly SmolVM) vs Vercel Sandbox
- Cloudflare Sandbox SDK vs Maritime
- Daytona vs Maritime
- Deno Sandbox vs Maritime
- E2B vs Maritime
- Freestyle vs Maritime
- InstaVM vs Maritime
- Maritime vs Microsoft Execution Containers
- Maritime vs Modal Sandboxes
- Maritime vs Morph Cloud
- Maritime vs Offscreen
- Maritime vs SPACE
- Maritime vs Runloop Devboxes
- Maritime vs Sandbox0
- Maritime vs Sprites
- Maritime vs Together Code Sandbox
- Maritime vs Vercel Sandbox
- Agent 37 Cloud vs Celesto (formerly SmolVM)
- Agent 37 Cloud vs Maritime
- DeltaBox Lite vs Maritime
Machine-readable
- This page as Markdown
/compare/celesto-vs-maritime.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/celesto.json·/api/v1/tools/maritime.json - From a terminal
anchor compare celesto maritime(the CLI) - Over MCP
compare_tools {"a": "celesto", "b": "maritime"}at/mcp, no key