Agentic Index
Fly.io Sprites vs Freestyle (2026)
Both vendors last verified
Both sell persistent Linux VMs for agents that pause when idle and wake on incoming traffic, and they differ in what a pause keeps and in the controls around each VM. That verdict is the Agentic Index coverage score, graded from each vendor's own published materials.
Fly.io Sprites pause about 30 seconds after activity stops, keep memory only through a warm pause, checkpoint the filesystem in about a second, and reach more than 40 services through connectors that hold credentials at Fly.io's gateway, with a hosted MCP server. Freestyle pauses VMs with memory and processes intact, clones running VMs, supports nested virtualization, starts every VM with its network closed, and injects API keys, database passwords and S3 signatures at its edge, with revocable client tokens for end users. Choose Sprites for computers that idle at no compute cost with ready connectors and an MCP server, and Freestyle when pauses must keep memory and you want tight control of each VM's network.
This comparison is published by Agentic Index, an independent agentic AI vendor research platform. Fly.io Sprites and Freestyle are each graded against the same 14 capability Agentic Index taxonomy, from the vendor's own public materials under the Agentic Index verification standard, alongside 1,165 researched vendors. No vendor pays for placement and no vendor has reviewed this page. How this evidence is graded
Choose Fly.io Sprites if
- Outside services such as GitHub, Slack or Notion should be reached through ready connectors.
- A hosted MCP server should let any MCP client create Sprites and run commands.
- Agents should checkpoint and restore their own filesystem, with recent checkpoints mounted read only.
- SDKs should cover Python, JavaScript, Go and Elixir.
- Your organization already uses Fly.io with Google or GitHub single sign on.
Choose Freestyle if
- A paused VM should keep its memory and running processes while it sleeps.
- Running VMs should be cloned with their memory.
- VMs should run Docker or nested VMs, with FUSE and eBPF.
- Database passwords and S3 credentials should stay at the edge while the VM connects.
- End users should get revocable tokens for their own VMs and Linux users.
| Feature | F Fly.io Sprites |
F Freestyle |
|---|---|---|
| Action & orchestration | ||
|
Integrations & Tool Calling Ability to connect agents to real systems through native integrations, OAuth-authenticated actions, custom tools, APIs, webhooks, or MCP-compatible tools. |
||
|
Fly.io SpritesIntegrations & Tool Calling Agents call outside services through connectors without holding the credentials. An organization stores a credential once, by OAuth for services such as GitHub or with a key, and the Sprite reaches the service through Fly.io's gateway, which identifies the calling Sprite, checks the connector's access policy and attaches the credential. The catalog holds more than 40 connectors, including GitHub, Slack, Notion, OpenAI, Anthropic, OpenRouter, Supabase, Neon, Composio and S3 compatible storage, a Custom API connector covers any token authenticated HTTP API, and a connector refuses every Sprite until a policy grants access by name, label or endpoint. Sourcedocs.fly.io/sprites/concepts/connectorsread 2026-11-05 |
||
|
FreestyleIntegrations & Tool Calling Credentials reach outside services without entering the VM. Outbound TLS rules let Freestyle's edge add API keys to requests for services such as OpenAI, Anthropic or OpenRouter, connect to external Postgres with a password the VM never sees, sign S3 requests limited to a bucket, prefix and set of operations, route traffic through the customer's SOCKS5 proxy and reach other VMs by internal names, and WireGuard tunnels connect VMs to remote networks. There are no prebuilt connectors or tool catalog for an agent to call. Sourcefreestyle.sh/docs/vms/network/tls-outboundread 2026-11-05 |
||
|
Workflow Orchestration Ability to sequence, branch, retry, route, and combine deterministic workflow nodes with autonomous agent steps. |
||
|
Fly.io SpritesWorkflow Orchestration There is no workflow logic for an agent's steps. Services keep long running processes up, restart them when they crash and start them in dependency order, and a Tasks API keeps a Sprite awake until work finishes, but sequencing, branching and retrying an agent's work stay in the customer's code or framework. Sourcedocs.fly.io/sprites/concepts/servicesread 2026-11-05 |
||
|
FreestyleWorkflow Orchestration There is no workflow logic in the product. Run limits pause a VM after a set time, TTLs delete it and snapshots support branching and rollback, but sequencing, retrying and routing an agent's steps stay in the customer's own code. Sourcefreestyle.sh/docs/vms/lifecycleread 2026-11-05 |
||
|
Triggers & Channel Coverage How agents wake up and where they work: schedules, webhooks, message events, CRM events, inbox events, chat, email, voice, and collaboration tools. |
||
|
Fly.io SpritesTriggers & Channel Coverage A request to a Sprite's URL wakes it without the customer's code calling it. If the Sprite's HTTP service is not running when the request arrives, the proxy starts it and forwards the request, so an agent or app served from a Sprite answers traffic while costing nothing in compute between requests. There is no scheduler or webhook. Sourcedocs.fly.io/sprites/concepts/servicesread 2026-11-05 |
||
|
FreestyleTriggers & Channel Coverage A paused VM wakes without the customer's code calling it. The next traffic that reaches the VM resumes it with its processes and memory intact, after a forward auth check when one is set, so an agent or app served from a Freestyle VM answers requests at its domain and pauses again when idle. There is no scheduler or webhook. Sourcefreestyle.sh/docs/vms/lifecycleread 2026-11-05 |
||
| Knowledge & context | ||
|
Knowledge Grounding & RAG Ability to ground agent behavior in company data through document ingestion, retrieval, external knowledge APIs, semantic search, or RAG layers. |
||
|
Fly.io SpritesKnowledge Grounding & RAG No retrieval index or knowledge base over the customer's content comes with Sprites. Files an agent reads live on the Sprite's disk, and any retrieval runs in the customer's own code. Sourcedocs.sprites.devread 2026-11-05 |
||
|
FreestyleKnowledge Grounding & RAG 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, and any retrieval runs as the customer's own software inside the VM. Sourcefreestyle.sh/docs/vmsread 2026-11-05 |
||
|
Memory & State Persistence Ability to persist context across a run, conversation, workflow, user, team, or longer-term memory layer. |
||
|
Fly.io SpritesMemory & State Persistence An agent's working state carries across sessions as machine state. A Sprite's disk keeps files, packages, repositories and databases stored on disk through every sleep, a warm pause keeps memory and running processes, and checkpoints copy the filesystem in about a second, with automatic checkpoints in the background and the last five mounted read only. An agent can checkpoint and restore its own Sprite. This is machine state the developer manages, not a memory layer of stated scope that can be reviewed or deleted as memory. Sourcedocs.fly.io/sprites/concepts/checkpointsread 2026-11-05 |
||
|
FreestyleMemory & State Persistence An agent's working state carries across sessions as machine state. A paused VM keeps its memory and resumes the same processes, persistent VMs keep their disks between runs, and snapshots and live cloning create new VMs that match the original for branching and recovery. Files on a paused VM stay readable without resuming it. This is machine state the developer manages, not a memory layer of stated scope that can be reviewed or deleted as memory. Sourcefreestyle.sh/docs/vms/lifecycleread 2026-11-05 |
||
| Control & trust | ||
|
Human Oversight & Guardrails Approval steps, consent checkpoints, escalation rules, structured guardrails, policy constraints, and pause/resume controls. |
||
|
Fly.io SpritesHuman Oversight & Guardrails Work in a Sprite has no approval step, consent checkpoint or escalation to a person. Connector policies, the outbound allowlist and MCP tokens limited to Sprites with a set name prefix restrict what an agent can reach, and an MCP client may ask before destructive tools run, but nothing on Fly.io's side holds an action for review. Sourcedocs.fly.io/sprites/integrations/remote-mcpread 2026-11-05 |
||
|
FreestyleHuman Oversight & Guardrails Work in a VM has no approval step, consent checkpoint or escalation to a person. A default closed firewall, forward auth on domains and credentials held at the edge limit what an agent can reach, but nothing holds an action until someone reviews it. Sourcefreestyle.sh/docs/vms/network/firewallread 2026-11-05 |
||
|
Security, Identity & Governance RBAC, SSO, auditability, encryption, least-privilege tool access, compliance posture, and data handling policy. |
||
|
Fly.io SpritesSecurity, Identity & Governance Sprites run in Fly.io organizations, which have admin and member roles and can require single sign on through Google or GitHub. Organization tokens authenticate the command line tool, the API and the SDKs, MCP access is granted by OAuth with tokens limited by default to Sprites with a set name prefix, Sprite URLs are private to the organization unless made public, and connector credentials are stored encrypted outside the machine. Fly.io offers SOC 2 Type II and a HIPAA package with a BAA signed in advance on its paid plans, without saying whether they extend to Sprites. Sourcedocs.fly.io/security/ssoread 2026-11-05 |
||
|
FreestyleSecurity, Identity & Governance Team API keys manage resources, and client sessions give an end user or agent its own identity with access to chosen VMs and Linux users through a token that can be revoked without rotating the team key. Secrets in outbound rules are write only and sealed at rest, and every VM starts with no network access until rules allow it. SAML single sign on and SCIM come with Enterprise, and Freestyle names no compliance attestation. Sourcefreestyle.sh/docs/vms/client-sessionsread 2026-11-05 |
||
|
Observability & Auditability Traces, logs, execution histories, metrics, audit events, and debugging detail for production agent behavior. |
||
|
Fly.io SpritesObservability & Auditability The machine is measured and the agent's steps are not. Service logs are kept with timestamps and read through the API, the command line tool and the MCP server, usage endpoints break compute and storage down by organization and by Sprite, managed connectors show usage over 24 hours, 7 days or 30 days, and exec sessions can be reattached after a disconnect. There are no traces or audit trail of what an agent did. Sourceapi.sprites.dev/openapi.jsonread 2026-11-05 |
||
|
FreestyleObservability & Auditability Audit logs come with the Enterprise plan. No logs, metrics or traces come with the VMs, so what an agent did inside a VM is visible only through the customer's own tooling and the command output returned to the caller. Sourcefreestyle.sh/pricingread 2026-11-05 |
||
|
Deployment & Data Residency Deployment modes and options, including SaaS, dedicated cloud, VPC, on-prem, hybrid, local runtime, and self-hosting. |
||
|
Fly.io SpritesDeployment & Data Residency Sprites run on Fly.io's hosted platform with no choice of region and no self hosted option. Creating a Sprite takes a name and URL settings only. Sourceapi.sprites.dev/openapi.jsonread 2026-11-05 |
||
|
FreestyleDeployment & Data Residency Freestyle runs as a hosted service with no choice of region and no self hosted or customer cloud option. VPCs and WireGuard tunnels connect VMs to the customer's networks without moving where the VMs run. Sourcefreestyle.sh/docs/vms/pricing-and-limitsread 2026-11-05 |
||
| Solution readiness | ||
|
Prebuilt Agents, Templates & Packs Ready-made workflows, packaged employees, templates, blueprints, industry solutions, and role-specific agents that reduce time-to-value. |
||
|
Fly.io SpritesPrebuilt Agents, Templates & Packs Starting points come as a ready image, plugins and framework integrations. Each Sprite starts from Ubuntu with Node.js, Python, Go, Rust, Java and other languages and AI command line tools preinstalled, plugins connect Claude Code, Codex, Cursor, OpenCode and other coding agents, and integrations cover Google ADK, the OpenAI Agents SDK, Pydantic AI and TanStack AI. These are tools a developer builds with, not prebuilt agents a business adopts. Sourcefly.io/sprites/ecosystemread 2026-11-05 |
||
|
FreestylePrebuilt Agents, Templates & Packs Starting points come as base snapshots and setup guides. Public base snapshots include Kali Linux and BusyBox and sit outside each plan's snapshot limit, and guides set up Claude Code, Python, Node.js, Go, Bun, Chromium, Docker, PostgreSQL, Redis, Supabase and Vite on a VM. These are environments a developer builds from, not prebuilt agents a business adopts. Sourcefreestyle.sh/docs/vms/base-snapshotsread 2026-11-05 |
||
| Platform extensibility | ||
|
Model Flexibility & Routing Ability to work across multiple foundation models, route tasks to different models, or let buyers bring their own providers and keys. |
||
|
Fly.io SpritesModel Flexibility & Routing The customer chooses the agents and models. Any coding agent or framework can run in a Sprite, and connectors reach OpenAI, Anthropic, OpenRouter, Fireworks, Hugging Face, Ollama Cloud and other model providers with the organization's own keys, or through a managed option for OpenRouter. Sourcefly.io/sprites/ecosystemread 2026-11-05 |
||
|
FreestyleModel Flexibility & Routing The customer chooses the agents and models that run on a VM. Outbound rules inject keys for providers such as OpenAI, Anthropic or OpenRouter, and a JSON patch rule can set or change the model field in request bodies at the edge, so the customer can pin which model an agent calls. Sourcefreestyle.sh/docs/vms/network/tls-outboundread 2026-11-05 |
||
|
APIs, SDKs & MCP Extensibility Composability layer: stable APIs, SDKs, MCP tool consumption/serving, custom tools, and integration into internal systems. |
||
|
Fly.io SpritesAPIs, SDKs & MCP Extensibility Sprites are managed through a REST API with a public OpenAPI specification, Python, JavaScript, Go and Elixir SDKs and the sprite command line tool. A hosted MCP server at sprites.dev/mcp, authorized by OAuth, enumerates, creates and destroys Sprites and runs commands, checkpoints, network policy, services and service logs inside them, and an agent inside a Sprite reaches the same controls through a local management socket. Sourcedocs.fly.io/sprites/integrations/remote-mcpread 2026-11-05 |
||
|
FreestyleAPIs, SDKs & MCP Extensibility A TypeScript SDK and the freestyle command line tool cover VMs, execution, files, snapshots, PTY sessions, domains, firewall rules, VPCs, tunnels, outbound TLS rules, client session identities and API tokens. Freestyle offers no MCP server or public REST reference. Sourcefreestyle.sh/docs/quickstartread 2026-11-05 |
||
|
Testing, Debugging & Optimization Testing, debugging, scoring, retries, fallbacks, quality gates, and optimization loops for improving agent workflows before and after deployment. |
||
|
Fly.io SpritesTesting, Debugging & Optimization There are no test cases, scoring or quality gates for an agent's behavior. Sprites can host evaluation and training environments such as Hugging Face OpenEnv through an experimental provider, with the scoring in the customer's own harness. Sourcedocs.fly.io/sprites/integrations/agent-sdksread 2026-11-05 |
||
|
FreestyleTesting, Debugging & Optimization There are no test cases, scoring or quality gates for an agent's behavior. Snapshots and cloning let experiments branch and roll back, with any evaluation done in the customer's own code. Sourcefreestyle.sh/docs/vmsread 2026-11-05 |
||
| Specialist automation | ||
|
Browser & Computer Use Browser, desktop, or remote/local computer control for workflows that cannot be handled through stable APIs alone. |
||
|
Fly.io SpritesBrowser & Computer Use No browser or desktop for an agent to drive comes with Sprites. The Steel Browser Tools connector returns scraped web content, screenshots and PDFs without browser sessions, and an agent that needs to drive a browser installs and runs one itself inside the Sprite. Sourcefly.io/sprites/ecosystemread 2026-11-05 |
||
|
FreestyleBrowser & Computer Use No browser or desktop for an agent to drive comes with the platform. Chromium appears among the setup guides for software to install on a VM, and an agent that needs a browser runs and drives it itself. Sourcefreestyle.shread 2026-11-05 |
||
Pricing snapshot
Sourced from the Index pricing dataset · open each vendor's profile for full detail.
| Pricing | F Fly.io Sprites |
F Freestyle |
|---|---|---|
|
Entry price Lowest public entry point |
Pay as you go at $0.0385 per CPU hour and $0.021875 per GB of memory an hour while a Sprite is awake, with $30 in trial credits for new users. | Hobby at $50 a month, which covers usage up to $50 before further usage bills, after a free plan with hard limits. |
|
Pricing confidence How public the numbers are |
Public, exact | Public, exact |
|
Billing Primary billing axis |
CPU hours and GB of memory hours while awake, plus hot storage while awake and cold storage until deleted | vCPU hours, GiB of memory hours and GiB of storage hours, plus data transfer |
|
Variable cost Workload / overage exposure |
High variable cost | High variable cost |
|
Free tier / trial Try before you buy |
No free tierTrial
|
Free tier
|
|
Buying motion Self-serve vs sales call |
Self-serve | Mixed |
|
Pricing verified When this pricing was last checked |
Sprites are a product of Fly.io, which offers SOC 2 Type II and a HIPAA package on its paid plans without saying whether they extend to Sprites. Freestyle offers no choice of region and has no public compliance attestation.
More comparisons with Fly.io Sprites or Freestyle
Other matchups in agent infrastructure platforms
Not the pairing you were after? These compare a different set of agent infrastructure platforms on the same 14 capabilities.