Code & AI agent sandboxes: read the guide

Freestyle

HostedSandbox

Full Linux VMs for AI Agents — Built for complex tasks that run for hours, days, or weeks.

Homepage Docs Pricing

Scanned 12 Oct 2026 · Schema v5 · features v21

From: $50/mo
Usage-based with a monthly plan fee that acts as a minimum spend credited toward usage. Plans: Free $0 / Hobby $50 / Pro $500 / Enterprise (custom). Paused (hibernated) VMs incur no compute charge.
Free tier: Not documented
Monthly hard allowances: 200 vCPU-hours, 400 GiB-memory-hours, 60000 GiB-storage-hours, 50 GB bidirectional transfer. 10 concurrent/saved VMs and snapshots. Free persistence disputed between pricing and docs.
Isolation: Virtual machine
hypervisor, or nothing at all
Regions: —
No workload region placement disclosed; no DNS/cloud-provider inference.
Capabilities: 25 / 61
3 partial · 33 fail · 0 not assessed
Agent readiness: 9/11 · 6.5/11
read · drive, pre-login surface

Advertised for

Long running AI agents · Untrusted code execution · Browser and desktop automation · Development environments · Evaluations · Nested containers and VMs

Built for

Developers, then technical

· SDK-first: `freestyle.vms.create()/fork()/snapshot()` TypeScript
· 'give your agents a sandbox VM' — for builders of AI products
· REST API + OpenAPI + CLI + SSH

01 Pricing Usage-based with a monthly plan fee that acts as a minimum spend credited toward usage. Plans: Free $0 / Hobby $50 / Pro $500 / Enterprise (custom). Paused (hibernated) VMs incur no compute charge.

Tier	EntryFree	Hobby	Pro	TopEnterprise
Price	$0/forever	$50/mo + usage over $50	$500/mo + usage over $500	custom
Includes	Concurrent 10; saved 10; snapshots 10; CPU 4; memory 8 GiB; disk 32 GB. Monthly included 200 vCPU-hours, 400 memory-GiB-hours, 60000 storage-GiB-hours.	Concurrent 40; saved 200; snapshots 1000; CPU 8; memory 16 GiB; disk 64 GB. Monthly included 200 vCPU-hours, 400 memory-GiB-hours, 60000 storage-GiB-hours.	Concurrent 400; saved 4000; snapshots 12000; CPU 32; memory 64 GiB; disk 256 GB. Monthly included 200 vCPU-hours, 400 memory-GiB-hours, 60000 storage-GiB-hours.	SAML SSO & SCIM, audit logs, 99.99% SLA, dedicated support, custom usage limits
Transfer	50 GB	500 GB	500 GB	—
vCPU hours	200 h	200 h	200 h	—
Memory gb hours	400 h	400 h	400 h	—
Transfer overage	—	$0.02/GB	$0.02/GB	—
vCPU hours overage	—	$0.04032/h	$0.04032/h	—
Memory gb hours overage	—	$0.0129/h	$0.0129/h	—

Free tier: Monthly hard allowances: 200 vCPU-hours, 400 GiB-memory-hours, 60000 GiB-storage-hours, 50 GB bidirectional transfer. 10 concurrent/saved VMs and snapshots. Free persistence disputed between pricing and docs.

Allocated vCPU $0.04032/hour + GiB memory $0.0129/hour + GiB storage $0.000086/hour + transfer $0.02/GB (both directions); compute metered per second.·Card required: unconfirmed·freestyle.sh

Overages: Monthly included compute/storage same on all plans; transfer 50 GB Free, 500 GB paid. Paid fees cover overage dollar-for-dollar. Paused VMs bill storage only; Free hard stops rather than overage.

02 Runtime Each fact carries its source; dots show confidence. Open a row for the evidence.

Isolation & lifecycle

Isolation primitive

Virtual machinehypervisor, or nothing at all

Persistence

Persistent saved disk; paused VMs retain memory and disk. Hobby/Pro persistence confirmed; Free differs between pricing table and docs.

Snapshots / forks

Live cloning includes memory; snapshots preserve running state. Homepage advertises 65 ms startup; API description says p99 under 400 ms; unbenchmarked vendor claims.

Resources & limits

Storage

VM root disk default 32 GB; max Free 32 GB, Hobby 64 GB, Pro 256 GB; billed while paused. External S3 integrations are not first-party object storage.

Limits & quotas

Concurrent VMs 10/40/400; saved 10/200/4000; snapshots 10/1000/12000. Max CPU 4/8/32; memory 8/16/64 GiB; disk 32/64/256 GB. Aggregate disk budgets 320/3125/62500 GB. Default 4 CPU/8 GiB/32 GB.

GPU

No GPU offering or GPU rates found on current pricing/docs.

Access

Deploy interface

apisdkcli

SDK languages

TypeScript/JavaScript

Network & secrets

Egress controls

Default-deny network; required allow rules with VM/VPC/tunnel/CIDR/port/protocol matchers. TLS/SNI domain rules can grant only permitted origins; broad firewall grants are additive.

Secrets

TLS edge encrypts configured header credentials at rest and masks readback; HTTP, Postgres and S3 credential injection keeps real credentials out of guest. Not a general-purpose vault.

Bring your own cloud

Placement & use

Jurisdiction

unknown

Advertised workloads

Long running AI agentsUntrusted code executionBrowser and desktop automationDevelopment environmentsEvaluationsNested containers and VMs

03 Capabilities 25 pass · 3 partial · 33 fail · 61 of 61 assessed

Supported — Documented on the public surface and verified.
Partially supported — Exists with caveats: gated, thin, or enterprise-only.
Not supported — Checked and not offered.
Not assessed — Scanned before this capability was tracked. Not a lack.

Compute & storage 6/11

Persistent diskStorage that survives restart/stop (not ephemeral-only).

Supported

Object storageManaged S3-compatible / blob object store the platform PROVIDES to a user's workloads (buckets, presigned URLs). DISTINCT from persistent-disk (block/volume attached to an instance) and from a self-host dependency (source.self_host.deps[]=["object-storage"] is an infra dep to RUN an OSS tool — the opposite direction). SCOPE varies and matters: native first-class (Railway Buckets, Cloudflare R2) vs a sibling SKU / extension (Fly Tigris, Bunny Storage, DigitalOcean Spaces) — record which in the evidence quote / note. Admitted v15.

Not supported

Snapshots / forkingSnapshot or copy-on-write fork of a running instance (filesystem-only counts). Whether the snapshot also restores memory + running processes is recorded separately by memory-snapshot (v19).

Supported

Live-state (memory) snapshotsA snapshot / checkpoint / pause captures a running instance's MEMORY and PROCESS state and restores it, so running processes continue where they left off instead of a freshly booted instance with the same files. Counts whether user-triggered (snapshot, checkpoint, pause-to-storage) or used to restore pre-initialised state (memory snapshots for fast cold starts). Distinct from snapshots-forks (any snapshot or fork, including filesystem-only; a product passing that key may still fail this one) and from scale-to-zero (an idle-cost property; suspension may or may not keep memory). Actor multiplexing (see scale-to-zero) records its live-state half here — only when the suspended actor's MEMORY/process state is what comes back, not a freshly booted worker with the actor's files.

Supported

Scale to zeroIdle instances suspend at no / negligible cost. ACTOR MULTIPLEXING (many suspended stateful actors sharing a small pool of pre-warmed workers, woken on request — Agent Substrate, AX, kagent; Rivet's hibernating actors) has NO key of its own (proposal REJECTED v20) — score the idle-cost half HERE and the live-state half under memory-snapshot; the oversubscription ratio / density is a performance characteristic the registry does not key.

Supported

GPU computeGPU-backed instances available.

Not supported

Multi-regionCompute placeable in more than one geographic region.

Not supported

Sandbox templates / custom imagesBuild reusable sandbox/VM images from an OCI image or Dockerfile (a base you fork instances from).

Supported

Cloud-init / ignition configCustomise a VM / microVM / instance at boot via cloud-init (or ignition) user-data — the same boot-metadata interface as AWS/Azure/GCP. Inject packages, users, SSH keys, files, and run-commands declaratively at launch. Distinct from sandbox-templates (a PREBUILT image you fork instances from) — this is PER-LAUNCH declarative customisation, not a baked image.

Not supported

Desktop / computer-use sandboxA virtual desktop (GUI) for LLM computer-use — browser + desktop automation, often via noVNC/VNC.

Supported

Selectable isolation runtimeThe user chooses the isolation runtime per workload / environment / sandbox from more than one option (e.g. microVM vs gVisor vs plain container, native namespaces vs gVisor, gVisor vs VM). Records the CHOICE; the isolation_primitive facet still records the best-evidenced primary. A platform that switches runtime automatically without a user choice (e.g. gVisor forced for GPU) does not pass on that alone.

Not supported

Networking & delivery 6/9

Automatic HTTPSTLS endpoints provisioned automatically per workload.

Supported

Built-in reverse proxyPlatform routes external traffic to workload services.

Supported

Custom domainsBring-your-own domain via CNAME/DNS.

Supported

Public preview URLsShareable per-deploy / per-branch URLs.

Supported

Direct public IPv6Per-instance stable public IPv6 for raw TCP/UDP (beyond a managed HTTPS proxy) — e.g. mosh / raw sockets.

Not supported

Egress controlsConfigurable OUTBOUND network policy — egress firewall / allowlist / egress proxy on traffic LEAVING a workload. Scoped to OUTBOUND only (v15): private-network REACHABILITY between a user's own services is private-networking, not this. Static-egress-IP belongs here (an outbound concept).

Supported

Private networkingA platform-internal private network between a user's OWN services — service-to-service connectivity, VPC, service mesh, and/or service discovery, not exposed to the public internet. DISTINCT from egress-controls (OUTBOUND firewall/allowlist), reverse-proxy (PUBLIC inbound ingress), and direct-ipv6 (per-instance PUBLIC addressing). Admitted v15.

Supported

Email send / receiveSend and/or receive email from the workload.

Not supported

Edge KV storeA globally-replicated key-value store read from WITHIN an edge runtime at the PoP (low-latency, eventually-consistent) — Cloudflare Workers KV, Deno KV, Fastly KV Store, Akamai EdgeKV, Vercel Edge Config. Distinct from first-party-db (a managed DATABASE service you provision and connect to) and from http-query-api (HTTP access to an external store) — edge-kv is colocated with the edge compute and read at the edge.

Not supported

Developer access 6/10

Public APIDocumented programmatic API.

Supported

OpenAPI specMachine-readable OpenAPI / Swagger spec published.

Supported

CLIOfficial command-line tool.

Supported

SDKOfficial client library/libraries.

Supported

SSH accessDirect SSH into workloads.

Supported

Filesystem APIRemote file operations over the API (list/read/write/watch/copy/upload) without SSH.

Supported

Git deployDeploy from a Git repo / push-to-deploy.

Not supported

Private container registryHosted private image registry.

Not supported

E2B-compatible APIDrop-in compatibility with the E2B sandbox API/SDKs — existing E2B client code runs against this product by changing the endpoint/domain + key, with no business-code changes. E2B itself passes as the reference implementation. Distinct from public-api / sdk (the product has its own API — this key says it also speaks another vendor's) and from self-host-oss (a self-hosted E2B-compatible server is scored on both).

Not supported

HTTP data APIQuery / read / write the stored DATA over plain HTTP/REST (or GraphQL) WITHOUT the native protocol driver (Postgres wire / Redis RESP) — agent- and edge-friendly data access via fetch(). The DATA plane; distinct from public-api (a control-plane / management API). Auto-generated-from-schema REST/GraphQL (PostgREST, pg_graphql) is a flavour of this.

Not supported

AI & agents 4/12

Built-in coding agentShips a first-party AI coding agent in the product.

Not supported

Agent skillsInstallable / declarable agent skills or tools.

Supported

AI app builderPrompt-to-app / vibe-coding generation surface.

Not supported

MCP serverExposes a Model Context Protocol server for agents.

Not supported

MCP clientActs as an MCP client / consumes MCP tools.

Not supported

MCP-server hostingRun/host MCP servers as a workload on the platform (distinct from exposing a first-party MCP control server — that is mcp-server).

Not supported

LLM gatewayRoutes LLM API calls through the platform (multi-provider).

Partially supported

BYOK for LLMsBring-your-own LLM provider keys.

Supported

Model inference API (first-party)A FIRST-PARTY hosted LLM/model inference API — the provider serves its OWN catalogue of (usually open-weight) models behind a (typically OpenAI-compatible) endpoint, billed per token. Distinct from llm-gateway (a multi-PROVIDER proxy/router for YOUR calls to external providers like OpenAI/Anthropic) and from byok-llm (bring your own provider keys) — here the platform IS the inference provider, serving its own models.

Not supported

Autonomous QA agentA first-party AUTONOMOUS AI agent that QA-TESTS a running app / implementation — exercising it like a real user (clicking through flows, checking behaviour) and reporting bugs/defects (often back to a coding agent, or before a human review). Distinct from built-in-agent (a code-AUTHORING agent — this one VERIFIES the result, it does not write the feature; a product can ship both), from ai-guided-code-review (organises a diff for a HUMAN to review — not an autonomous tester), and from observability (runtime metrics/logs, not active testing).

Not supported

Credential-injection proxyThe platform holds real secrets OUTSIDE the workload and injects them on the wire (an egress / broker proxy swaps a placeholder or phantom token for the real credential on matching outbound requests), so code running in the sandbox — e.g. an untrusted agent — never sees or can exfiltrate the real key. Distinct from secrets-management (storage + injection INTO the workload as env vars/files — the workload holds the secret) and from egress-controls (outbound allow/deny POLICY; a product can have an allowlist without injecting credentials, and vice versa).

Supported

Preinstalled coding agentThe platform installs and launches a THIRD-PARTY coding agent (a harnesses.yml entry — Claude Code, Codex, OpenCode, Pi ...) inside a workspace/sandbox it provisions, with the agent's auth wired in (your own account or API key), so you start an agent session rather than an empty box. Distinct from built-in-agent (a FIRST-PARTY agent the vendor ships) and from harnesses[] (which records that a harness CAN target the product, not that the product runs it for you).

Supported

Data 0/4

First-party databasePlatform-provided managed database.

Not supported

Database branchingCopy-on-write branch of a DATABASE (schema + data) as a dev-lifecycle primitive — branch your data like code, usually instant via a metadata operation. Distinct from snapshots-forks (forks a running compute INSTANCE / VM state) and from preview-environments (a full per-PR environment that may USE a DB branch). For a data product a CoW branch maps HERE, not to snapshots-forks.

Not supported

Vector searchFirst-class storage, indexing, and similarity search of vector embeddings — a native vector type/index or a managed vector DB. The data-layer primitive for RAG / semantic retrieval.

Not supported

Ephemeral instant DB (no-signup)Provision a working, usable instance via a single command / HTTP call with NO account, signup, or card — typically time-boxed (expires unless claimed). A zero-onboarding path an agent can hit directly.

Not supported

Platform & operations 0/6

Scheduled tasksCron / scheduled job primitive.

Not supported

Durable workflowsDurable execution / long-running workflow primitive.

Not supported

Preview environmentsA full EPHEMERAL environment spun up per PR / branch — compute + addons + (often) a DB branch — torn down on merge. More than preview-urls (which is just a shareable URL for one deploy).

Not supported

Built-in observabilityFirst-party metrics / logs / tracing / APM / profiling included in the platform (no third-party stack required to see what a workload is doing).

Not supported

Managed CI/CD runnersPlatform-hosted, drop-in CI/CD runner VMs (GitHub Actions / GitLab), typically faster and cheaper than the CI vendor's own hosted runners, billed per runner-minute; isolated + ephemeral per job. Distinct from git-deploy (push-to-deploy of an APP) and scheduled-tasks (cron) — this is managed CI COMPUTE you point your existing pipelines at.

Not supported

Project template marketplaceA marketplace / gallery of prebuilt PROJECT STARTER templates (apps or websites) a user clones or launches a new project from and then customises — the "start from a template" surface of a builder / PaaS. Distinct from sandbox-templates (a COMPUTE base IMAGE you fork runtime INSTANCES from — Dockerfile / OCI), from app-builder (the prompt-to-app GENERATION surface — a product can have both), and from one-click OSS-app catalogues / helm-addon-marketplace (deploy a FINISHED app as-is, not a starter you build on).

Not supported

Organisation & security 3/9

Built-in authPlatform-provided authentication / IAM for workloads.

Partially supported

SSO / SAMLEnterprise single sign-on.

Partially supported

Secrets managementFirst-class secrets storage / injection.

Supported

TeamsMulti-user team accounts.

Supported

SharingShare a running workload/resource via link (optionally gated).

Supported

BYOCRuns inside the customer's own cloud account.

Not supported

Deploy into customers' VPCsA SaaS vendor uses the platform to deploy and operate ITS software inside ITS CUSTOMERS' cloud accounts / VPCs (nested or "customer BYOC"), managed from the vendor's control plane. Distinct from byoc (your own workloads in your own cloud account).

Not supported

Workload identityThe platform gives each workload its own verifiable identity (OIDC token, SPIFFE SVID, cloud IAM federation) that it exchanges for short-lived credentials, so no long-lived static secret is stored or injected. Distinct from secrets-management (stores / injects static secrets) and credential-injection-proxy (swaps real secrets in on the wire).

Not supported

Self-host / OSS coreOpen-source core that can be self-hosted.

Not supported

04 Agent readiness Pre-login surface · counts, not a grade

Can an agent read it?

parse-ability · docs, metadata, machine-readable surfaces

9/11

7 full · 4 partial · 82% of applicable checks

llms.txt
Supported
“Rich /llms.txt (pages + all blog articles + full docs index with per-page descriptions + .md URLs) at both www and docs.”
llms-full.txt
Supported
“Docs llms-full.txt fetched (860331 bytes, pointer retained); comprehensive modular .md guides also saved.”
robots.txt AI policy
Partially supported
“robots.txt is `User-agent: * / Allow: /` — allows all bots but does not explicitly name AI UAs (silent default-allow).”
JSON-LD
Partially supported
“Homepage JSON-LD: Organization + WebSite + ItemList + BreadcrumbList — more than minimal, but no SoftwareApplication/Product/offers graph.”
Sitemap
Supported
“Sitemap index links sitemap-site.xml and docs/sitemap.xml; both fetched.”
Meta tags
Supported
“Homepage title Freestyle - VMs for AI Agents with descriptive metadata and OG tags.”
Markdown negotiation
Partially supported
“Clean per-doc .md variants exist; regular homepage fetch is HTML, no evidence of content-negotiated markdown.”
Content efficiency
Partially supported
“Homepage readable text / HTML bytes approximately 2.94%; measured after stripping scripts/styles and tags.”
Docs discoverable
Supported
“Docs link first-class in homepage header; canonical docs at www.freestyle.sh/docs.”
Machine-readable pricing
Supported
“Structured pricing page: per-resource rate tables + per-plan limit tables (Git tab JS-rendered/not captured).”
No anti-bot wall
Supported
“No challenge on public pages/docs; fetched cleanly.”

Can an agent drive it?

act-ability · APIs, SDKs, MCP, credentials, webhooks

6.5/11

5 full · 3 partial · 59% of applicable checks

Public API
Supported
“Current v5 REST API and VM/networking SDK docs.”
OpenAPI spec
Supported
“Fetchable OpenAPI 3.1.0 with 53 paths.”
MCP server
Not supported
“MCP well-known probes return 404; no first-party MCP control server setup found in current docs.”
SDK packages
Partially supported
“TypeScript SDK is primary/current; the Python SDK is a legacy Sandboxes-era package — effectively one current language.”
Auth documented
Supported
“CLI browser login, account API key minting and scoped per-identity VM tokens documented.”
Self-serve credentials
Supported
“Logged-in CLI documents tokens create --output json (token field); initial browser authentication remains required.”
Webhooks
Not supported
“No current platform webhook/signing API or docs found; former Git push triggers are absent from current spec.”
CLI tool
Supported
“Official `freestyle` CLI (npm).”
Error model
Partially supported
“API specifies per-endpoint error responses, but no unified public error taxonomy found.”
Rate limits documented
Not supported
“No rate-limit documentation found on the public surface.”
Agent config files
Partially supported
“Installable freestyle-docs skill and onboarding page; no fetched agents.md/well-known agent configuration.”

05 Coding harnesses Which agents and IDEs are documented to run here

Advertised integrations 6

Claude Code Anthropic
Supported
“Run Claude Code over PTY or Agent SDK.”
Source freestyle.sh
Codex CLI OpenAI
Supported
“Ubuntu includes codex.”
Source freestyle.sh
OpenCode opencode
Supported
“Ubuntu includes opencode.”
Source freestyle.sh
OpenClaw Peter Steinberger
Supported
“Run OpenClaw in a VM snapshot.”
Source freestyle.sh
GitHub Copilot GitHub
Supported
“Run GitHub Copilot CLI headless.”
Source freestyle.sh
Cursor Anysphere
Supported
“Cursor remote editor via SSH.”
Source freestyle.sh

Plausible compatibility 12

Derived, not stored: the product passes SSH access, so any CLI coding agent is installable inside it. This is plausible compatibility, not a supported integration.

Aider
Amp
Antigravity
Devin
Droid
Gemini CLI
Goose
Hermes Agent
Kilo Code
Kiro CLI
Pi
Qwen Code

06 Compliance & regions Standards held, with ordinal rigor; never a score

Certifications held 0

No attestations found on the public surface. Empty means nothing is advertised, after checking, not that nothing was checked.

Regions —

Regions are not disclosed on the public surface.

No workload region placement disclosed; no DNS/cloud-provider inference.

Legal

Privacy notice

07 Company & presence Facts only, not scored

Company

Headquarters: San Francisco, CA, USA
Founded: Unknown
Team size: Small early-stage startup; not disclosed.
Hiring: Hiring engineering roles, San Francisco, full-time.

Activity

Last release: Current public API v5; no dated platform release index found. api.freestyle.sh
Last blog post: Active blog index fetched; no reliable latest-post timestamp extracted. freestyle.sh

Presence 6

Links found on the public surface scanned; not an exhaustive search.

08 Sources & method How this entry was produced

Public surface only

Everything here was read from freestyle.sh's unauthenticated pages, docs, pricing and repository on 12 Oct 2026. Nothing behind a login was used.

Statuses, not scores

Capabilities are pass, partial or fail against a written rubric (features v21). “Not assessed” means the scan predates the key.

Confidence is separate

Each fact carries its own confidence. Low-confidence facts are shown, never hidden, so you can weigh them.

Schema: scan v5
Features registry: v21
Keys tracked today: 61
Assessed on this entry: 61

Something wrong on this page?

Nearby in the atlas

All Sandbox products →