How much does exe.dev cost?

Personal costs $15 a month for a pool of 2 vCPUs and 4 GiB with up to 50 VMs, with larger pools up to $155 a month. Work costs $75 a month plus $0.105 an hour for each extra 2 vCPU and 4 GiB unit, and Enterprise is priced by contract.

Does exe.dev support MCP and tool calling?

Full under the Agentic Index verification standard. exe.dev publishes public evidence of Model Context Protocol (MCP) support, tool calling, and API or SDK extensibility for connecting agents to external tools. Evidence basis: An HTTPS API runs the same command set as the SSH interface, covering VMs, sharing, resizing, tags, copies, pools, integrations and teams, with bearer tokens limited by expiry and allowed commands, and an explorer walks through each command. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Can exe.dev be self hosted or run in a private VPC?

Full under the Agentic Index verification standard. exe.dev publishes public evidence of self hosted or private VPC deployment, along with data residency options for controlling where agent data is processed and stored. Evidence basis: VMs run in a region chosen per account or per pool, among Los Angeles, New York, Dallas, Frankfurt, London, Tokyo, Singapore and Sydney, with Oregon closed to new accounts, while the main exe.dev server sits in the United States for everyon. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Can you bring your own model or route between providers with exe.dev?

Full under the Agentic Index verification standard. exe.dev publishes public evidence of bring your own model and bring your own key (BYOK), and of routing tasks between models or providers rather than being locked to one. Evidence basis: The customer chooses the models. The LLM integration routes requests by model ID to OpenAI, Anthropic, Fireworks or a custom HTTPS provider, using exe.dev managed credentials, the customer's own keys, a ChatGPT subscription or workload iden. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Back to vendors

exe.dev

Q: Who is exe.dev for?

developers, agent builders, engineering teams

Also known as: exe, Shelley, Bold Software, exe.xyz

Visit site

Agent infrastructureindependentVerified 2026-11-05

Entry pricePersonal at $15 a month for a pool of 2 vCPUs and 4 GiB of memory with up to 50 VMs, after a 7 day free trial.Full pricing detail

exe.dev sells persistent Linux VMs for agents that you reach over SSH, with HTTPS sharing, more than a hundred integrations whose secrets never reach the VM, an LLM gateway, a choice of eight regions and Shelley, a bundled web based coding agent.

exe.dev is a cloud of persistent Linux VMs for agents and the people who build with them, run by Bold Software, a small team in Berkeley, California. A customer buys a pool of CPU and memory and runs as many VMs in it as fit, each created in about a second over SSH with root access, a persistent NVMe disk and an HTTPS address at an exe.xyz subdomain. VMs stay up rather than scaling to zero, and can be copied, backed up, resized and shared with specific people, the whole team or the public.

Integrations connect a VM to other services while keeping the secrets on exe.dev's side. The credential is added to requests at the network edge, so the VM and any agent on it can use GitHub, Slack, Discord, S3 storage, Google Calendar and Gmail or more than a hundred other catalog services without being able to read the key.

An LLM gateway does the same for OpenAI, Anthropic, Fireworks and custom providers, with exe.dev managed credits or the customer's own keys, and an egress setting can limit a VM to its integrations alone. Each VM can receive and send email at its own address. The default image ships Shelley, exe.dev's web and mobile coding agent, which can run parallel conversations, compact long contexts and drive a Chromium browser with screenshots.

The Personal plan costs $15 a month for a pool of 2 vCPUs and 4 GiB with up to 50 VMs, the Work plan $75 a month with hourly billing for extra capacity at $0.105 per 2 vCPU and 4 GiB unit and unlimited users with single sign on, and Enterprise adds dedicated hardware and VMs in the customer's own cloud or data center. Teams have billing owner, admin and user roles and can enforce Google or OIDC sign in, and API tokens carry an expiry and a set of allowed commands. VMs run in Los Angeles, New York, Dallas, Frankfurt, London, Tokyo, Singapore or Sydney, with Oregon closed to new accounts. exe.dev has no public compliance attestation.

exe.dev fits developers and teams who want long lived computers for their agents and internal tools, reached by SSH, with integrations that keep credentials out of the agent's reach and a choice of region. It fits less well for buyers who want a compliance report, approval steps on what an agent does, or sandboxes that scale to zero and wake on demand.

Vendor details

Canonical URL

https://exe.dev

Funding status

exe.dev is run by Bold Software, Inc. of Berkeley, California, which has no disclosed funding.

Company status

independent

Use cases & customers

Primary use cases

agent sandboxescoding agentsinternal tools hostingpersistent dev environmentsemail and chat bots

Target customers

developersagent buildersengineering teams

Deployment options

SaaSAPIBring your own cloudSingle tenant hardware

Integrations

Integrations inject credentials at the network edge for GitHub, Slack and Slack bots, Discord and Discord bots, S3 compatible storage including Amazon S3, Tigris and Cloudflare R2, Google Calendar and Gmail, an LLM gateway for OpenAI, Anthropic, Fireworks and custom providers, VM to VM calls, tunnels to private networks and more than a hundred catalog services. Workload identity federation covers AWS, Google Cloud, Anthropic, OpenAI and Tailscale. VMs are managed over SSH, an HTTPS API that runs the same commands and an MCP server with plugins for Claude and ChatGPT, and VS Code, Claude Code and Codex run on the VMs alongside Shelley.

In practice

You want an agent to run on its own computer for weeks, not minutes. Create an exe.dev VM over SSH, start your agent on it and leave it running, with its files on a persistent disk and its web interface at a private HTTPS address only your team can open.

Your agent needs to commit to private GitHub repositories and call Anthropic's API, but must never see the keys. Attach the GitHub and LLM integrations to its VM and set egress to integrations only, so the agent can reach those services and nothing else on the internet.

You want an agent that works through the support requests people email to it. Turn on email receiving for its VM, have it watch the mailbox for new messages, and let it reply within each sender's thread.

Sources & related URLs

Related / legacy domains

https://github.com/boldsoftware/shelley

Research sources

https://exe.dev/ https://exe.dev/pricing https://exe.dev/docs https://exe.dev/docs/integrations https://exe.dev/docs/integrations-llm https://exe.dev/docs/shelley/intro https://exe.dev/docs/shelley/unique https://exe.dev/docs/teams/overview https://exe.dev/docs/regions https://exe.dev/docs/enterprise-byoc https://exe.dev/docs/https-api https://exe.dev/docs/receive-email https://exe.dev/docs/standalone-vms https://exe.dev/docs/terms-of-service https://exe.dev/security

Agentic Index coverage score

7.5 / 14 capabilities · 54%

Integrations & Tool Calling	Full
Agents use outside services through integrations whose secrets stay on exe.dev's servers and are added to requests at the network edge, so the VM can call the service but cannot read the key. Built in types cover GitHub, Slack and Slack bots, Discord and Discord bots, S3 compatible storage, LLM providers, calls between VMs and tunnels to private networks, the catalog covers more than a hundred services, Google Calendar and Gmail let an agent add events and send email as the user, and workload identity federation covers AWS, Google Cloud, Anthropic, OpenAI and Tailscale. Integrations attach to VMs directly, by tag or to every VM. Sourceexe.dev/docs/integrationsread 2026-11-05
Workflow Orchestration	Not documented
There is no workflow logic for an agent's steps. Shelley can run several conversations at once on one VM or across VMs, each its own session, and pools share capacity across VMs, but sequencing, branching and retrying work stay in the agent or the customer's code. Sourceexe.dev/docs/shelley/uniqueread 2026-11-05
Knowledge Grounding & RAG	Not documented
No retrieval index or knowledge base over the customer's content comes with the platform. Files an agent reads live on the VM's disk or in attached storage, and Shelley reads project instructions from an AGENTS.md file. Sourceexe.dev/docsread 2026-11-05
Human Oversight & Guardrails	Not documented
No approval step, consent checkpoint or escalation to a person comes with Shelley or with agents on a VM. Egress set to integrations only, private HTTPS addresses and secrets the VM cannot read limit what an agent can reach, but nothing holds an action until someone reviews it. Sourceexe.dev/docs/internet-accessread 2026-11-05
Security, Identity & Governance	Partial
Teams have a billing owner, admins who manage members and can reach any member's VM, and users who manage their own VMs, and a team can require sign in through Google or a custom OIDC provider such as Okta or Azure AD. API tokens carry an expiry and a set of allowed commands, SSH keys can be scoped to one VM, HTTPS addresses are private by default and integration secrets never reach the VM. exe.dev has no public compliance attestation, and its terms offer no certifications. Sourceexe.dev/docs/teams/overviewread 2026-11-05
Observability & Auditability	Partial
Each VM can be inspected with resource metrics drawn as sparklines, and Shelley's web interface keeps every step of each conversation in view as it runs. No audit log, log service or traces for other agents on a VM come with the platform. Sourceexe.dev/apiread 2026-11-05
Memory & State Persistence	Partial
An agent's working state lasts because the computer does. VMs are long lived with persistent NVMe disks whose changes are streamed to other machines in the region, and they can be backed up and copied. Shelley keeps each conversation as a log and compacts older context into new generations so long work can continue. This is machine and conversation state the user manages, not a memory layer of stated scope that can be reviewed or deleted as memory. Sourceexe.dev/docs/what-is-exeread 2026-11-05
Deployment & Data Residency	Full
VMs run in a region chosen per account or per pool, among Los Angeles, New York, Dallas, Frankfurt, London, Tokyo, Singapore and Sydney, with Oregon closed to new accounts, while the main exe.dev server sits in the United States for everyone. Enterprise customers can have dedicated single tenant hardware or run their team's VMs on servers in their own cloud account or data center. Sourceexe.dev/docs/regionsread 2026-11-05
Prebuilt Agents / Templates / Packs	Partial
The default VM image comes with Shelley, a general coding agent, already running, and VMs can start from custom Docker images. Repository buttons and deploy guides start projects on a VM. These are a starting agent and environments a developer works from, not a set of prebuilt agents for business tasks. Sourceexe.dev/docs/shelley/introread 2026-11-05
Triggers & Channel Coverage	Full
Work can reach an agent on its own. A VM can receive email at its own address, delivered to a mailbox the agent watches, and reply within each sender's thread, and Slack and Discord bot integrations carry messages both ways. VMs stay running, so an agent on one can act on incoming mail, chat messages or requests to its public HTTPS address as they arrive. There is no built in scheduler. Sourceexe.dev/docs/receive-emailread 2026-11-05
Model Flexibility & Routing	Full
The customer chooses the models. The LLM integration routes requests by model ID to OpenAI, Anthropic, Fireworks or a custom HTTPS provider, using exe.dev managed credentials, the customer's own keys, a ChatGPT subscription or workload identity, and Shelley offers whichever models the attached integrations provide. Claude Code, Codex and other agents on the VM can point at the same gateway. Sourceexe.dev/docs/integrations-llmread 2026-11-05
APIs / SDKs / MCP Extensibility	Full
An HTTPS API runs the same command set as the SSH interface, covering VMs, sharing, resizing, tags, copies, pools, integrations and teams, with bearer tokens limited by expiry and allowed commands, and an explorer walks through each command. An MCP server at exe.dev/mcp connects Claude Code and other agents, with plugins for Claude and ChatGPT, though its tools are not enumerated publicly. Sourceexe.dev/docs/https-apiread 2026-11-05
Testing, Debugging & Optimization	Not documented
There are no test cases, scoring or quality gates for an agent's behavior. Standalone VMs suit agent written code and CI jobs, with any testing done by the customer's own tools. Sourceexe.dev/docs/standalone-vmsread 2026-11-05
Browser / Computer-use	Partial
Shelley, the agent on the default image, has a browser tool that drives Chromium, navigating sites, taking screenshots and using browser profiles. Other agents on a VM get no browser of exe.dev's own and install and drive one themselves. Sourceexe.dev/docs/shelley/uniqueread 2026-11-05

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Pricing

Personal at $15 a month for a pool of 2 vCPUs and 4 GiB of memory with up to 50 VMs, after a 7 day free trial.

Monthly subscription for a shared pool of vCPU and memory, plus hourly metering for capacity beyond it

Trial available

Included quota

Personal includes a 2 vCPU, 4 GiB pool, up to 50 VMs, a 100 GiB filesystem and 200 GiB of network bandwidth, with larger pools of 4 vCPU and 8 GiB at $35, 8 vCPU and 16 GiB at $75 and 16 vCPU and 32 GiB at $155 a month. Work includes 2 vCPUs and 4 GiB across pools, up to 1,000 VMs per pool, unlimited pools in any region, a 100 GiB filesystem, 100 GiB of bandwidth and unlimited users with SSO.

What is public

exe.dev publishes both plan prices, the pool size options and prices, VM and bandwidth limits, the hourly rate for extra capacity and standalone VMs, and what Enterprise adds.

Billing mechanics

A subscription buys a pool of vCPU and memory shared by all VMs in it, with no per VM charge. Extra Work capacity and standalone VMs meter at $0.105 an hour per 2 vCPU and 4 GiB unit, billed by the second.

Cost watchouts

Standalone VMs bill by the second on top of every subscription except the Work plan's included capacity, and a stopped standalone VM still counts toward VM limits. The Work plan's $37.50 rate lasts only the first three months.

Variable cost rationale

The monthly pool caps what the always on VMs inside it can use, so cost stays fixed until a team adds capacity, standalone VMs or Work pool units, which meter by the hour.

Additional watchouts

A team on Work with 8 vCPUs and 16 GiB running all month pays the $75 base plus three extra units at about $0.105 an hour each, roughly $230 more a month.

Overage / add-ons

Capacity beyond a plan's pool, and standalone VMs on every plan, bill at $0.105 an hour per 2 vCPU and 4 GiB unit, metered by the second.

Sales call required

Mixed (some tiers require a call)

Free / trial

7 day free trial on the Personal plan

Lowest paid plan

Personal, $15 a month

Commercial notes

Bold Software, Inc. of Berkeley sells exe.dev self serve on Personal and Work, and Enterprise covers dedicated hardware, VPC support and VMs on the customer's own servers by contract.

Key ambiguities

exe.dev gives a 25 GB NVMe disk in one place and a 100 GiB filesystem per plan in another, and the Work plan's scale of 2 to 512 has no unit.

Missing data

Enterprise prices are not public, and managed LLM gateway usage has no rate beyond Deepgram transcription.

Agentic Index verified 2026-11-05

Official pricing page Docs

Data confidence: medium

Head to head

exe.dev vs boxdexe.dev vs Fly.io Sprites