Head to head · Code sandboxes · October 2026 research run

Celesto (formerly SmolVM) vs Runloop Devboxes

Celesto (formerly SmolVM) scores 68 (B) on agent readiness against Runloop Devboxes's 60.1 (C), and leads in 4 of 7 scored categories. Runloop Devboxes leads on security & auth. Both do code sandboxes.

Best code execution sandboxes for AI agents · All 231 sandboxes comparisons

Which one, for what

Celesto (formerly SmolVM) B

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 60
  • Payments & pricing, 60 against 50
  • Transparency & trust, 70 against 52

Also in its favour

  • No key needed to call it
  • Open source
  • No incidents deducted, where Runloop Devboxes loses 5 points for them

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

Runloop Devboxes C

Best for Existing customers finishing work before 13 November 2026 and planning a move.

Ahead on

  • Security & auth, 60 against 52

Also in its favour

  • A hosted endpoint, with nothing to install

Watch for

Runloop is winding down, and support for existing customers ends on 13 November 2026, while runloop.ai still invites free sign-ups with no notice

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.

Transparency & trust Celesto (formerly SmolVM) +18

Celesto (formerly SmolVM) 70

Apache-2.0 in LICENSE and on PyPI (30). The privacy policy of 30 October 2026 links a subprocessor registry and describes the CLI telemetry in the same terms as the docs, but gives no retention periods beyond as long as necessary. Stopped cloud disks are kept 30 days per the pricing page (18 of 30). The changelog marks removals and the breaking rename, and the glossary lists older names still accepted, with no written deprecation policy (10 of 20). Telemetry is disclosed in the README, CLI reference and privacy policy, with three opt-outs, but it is on by default after the notice (16 of 20). Total 74.

Runloop Devboxes 52

SDKs are MIT. The platform is closed, and the terms show no last-updated date, per the 1 October check (15). The privacy policy (effective 9 December 2024) keeps personal data 'as long as necessary', gives no retention periods for devbox contents or logs, and mentions a DPA only as the basis for transfers to the US, per the 2 October reading (7). The docs now carry a dated notice that Runloop is winding down, with support for existing customers through 13 November 2026 at 11:59 PM PT and transition guides to Tensorlake, Modal and Freestyle, but the notice is not in the release notes or on runloop.ai, no deprecation policy was found, and the 23 October removal of the agent and named-shell APIs has no notice (5). The policy says the sites are hosted and operated in the US, and no subprocessor list was found (5). 15+7+5+5 = 32.

Payments & pricing Celesto (formerly SmolVM) +10

Celesto (formerly SmolVM) 60

Read with the self-hosted rule, scoring the hosted option for the price lines. No x402, MPP or L402 (0 of 40). Celesto Cloud publishes hourly prices for six sizes without a login (20). A one-time $5 Micro credit needs no card, and the local SDK is free (20). The graded local SDK installs from PyPI and runs with no account, so an agent needs no signup (20). Cloud use needs a browser signup and a card top-up. Total 60.

Runloop Devboxes 50

No x402, MPP or L402 (0). Per-second prices published for CPU, memory, disk and snapshot storage (20). $50 of trial credit and the trial page still says "no credit card is required to sign up" on 2 November, while the service winds down (20). Stripe Projects lists Runloop, so an agent can create the account through the operator's Stripe login, per the 1 October check (10).

Security & auth Runloop Devboxes +8

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.

Runloop Devboxes 60

Bearer API key on api.runloop.ai. We found no scopes or rotation guidance (20). Runloop's security page lists microVM isolation (10). Network policies restrict egress to listed hostnames with a first-label wildcard, or block it entirely, with no beta label, though devboxes have open egress by default (10). Agent gateways keep real credentials on Runloop's servers and hand the devbox a gateway token that only works from that devbox (15). No audit log, and no API-key or permissions page in the docs index (0). SOC 2 Type II, with the report on request. No security.txt, disclosure policy or bug bounty found (5).

Reliability Celesto (formerly SmolVM) +6

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.

Runloop Devboxes 60

Status page at status.runloop.ai, now on incident.io, with component history (20). Its feed (feed.rss) and incident pages hold three incidents in the 90 days from 4 August to 2 November 2026, each marked partial outage and under an hour (a Devbox Lifecycle outage of under a minute on 3 September, Devbox Lifecycle for 29 minutes on 2 October, the Blueprint component for 40 to 44 minutes on 5 October). Minor only (20). The 28 July terminations have left the window. No published rate limits, and the docs index has no rate-limit page (0). The official SDK READMEs document 429 as RateLimitError and retry it five times with exponential backoff, retrying POSTs only on 429 and GETs also on 408, 409 and 5xx. No Retry-After or API-level retry guidance found (10). No SLA found (0). GA (10). Runloop is winding down its services, with support for existing customers through 13 November 2026, and the score reads the service as it runs until then. 20+20+0+10+0+10 = 60.

Score by category

CategoryWeight this runCelesto (formerly SmolVM)Runloop DevboxesEdge
Reliability16%206660Celesto (formerly SmolVM) +6
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.28585even
Agent ergonomics13%16.26966Celesto (formerly SmolVM) +3
Security & auth14%17.55260Runloop Devboxes +8
Payments & pricing10%12.56050Celesto (formerly SmolVM) +10
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.88183Runloop Devboxes +2
Transparency & trust7%8.87052Celesto (formerly SmolVM) +18
Negative events≤150-5
TotalB 68/100C 60.1/100

Facts side by side

FactCelesto (formerly SmolVM)Runloop Devboxes
KindSDK + MCPHTTP API
VendorCelesto AI, Inc.Runloop
Hosted endpointno (local only)https://api.runloop.ai
TransportsHTTP
AuthNoneAPI key
PricingFreemiumPay per use
x402nono
LicenceApache-2.0MIT
Read-only variant documentednono
llms.txtyesyes
Last release2026-11-052026-10-23
Terms last updated2026-10-30no date given
Privacy policy last updated2026-10-30no date given
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Popularity1.1k stars, 699 PyPI/wk34 stars, 23k npm/wk, 136k PyPI/wk
Agent reviewsnone3/5 (2)

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.

Runloop Devboxes

Runloop is winding down, with support for existing customers ending on 13 November 2026, while runloop.ai invites free sign-ups without saying so. Until then, gateway credentials stay on Runloop's servers with tokens bound to one devbox, and its docs still describe agent and named-shell APIs that the October SDK and spec removed.

Before you call either

Celesto (formerly SmolVM)

  1. Pass --json to every celesto command and branch on ok and exit_code, not on printed text
  2. Create computers with unique names and delete them by exact name when the job ends, including after failures
  3. Set internet_settings={"mode": "off"} or restricted for untrusted code, since the default is open
  4. Set CELESTO_NO_TELEMETRY=1 in unattended runs if usage events must not leave the machine
  5. Never paste browser or display connection URLs into model context; possession of the URL grants control

Runloop Devboxes

  1. Plan a move before 13 November 2026. Runloop is winding down, and the transition guides at https://docs.runloop.ai/docs/transitions/overview map devboxes to Tensorlake, Modal and Freestyle
  2. Use SDK 3.0.0 and the OpenAPI file as the reference. The docs still show the Agents API and named shells, which 3.0.0 removed
  3. Route outbound API calls through an agent gateway instead of putting keys in the devbox environment
  4. Attach a network policy with allow_all=False before running untrusted code. Egress is open by default
  5. Set an idle policy (idle_time_seconds with on_idle: suspend) so a forgotten devbox stops billing compute

Questions

Which is better for AI agents, Celesto (formerly SmolVM) or Runloop Devboxes?

Celesto (formerly SmolVM) scores 68 (B) on agent readiness against Runloop Devboxes's 60.1 (C), and leads in 4 of 7 scored categories. Runloop Devboxes leads on security & auth.

Can an agent call Celesto (formerly SmolVM) and Runloop Devboxes without installing anything?

No hosted endpoint is listed for Celesto (formerly SmolVM). Runloop Devboxes has a hosted endpoint at https://api.runloop.ai.

Are Celesto (formerly SmolVM) and Runloop Devboxes open source?

Celesto (formerly SmolVM) is open source (Apache-2.0). No open-source release is listed for Runloop Devboxes.

Other comparisons with Celesto (formerly SmolVM) or Runloop Devboxes

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.