Daytona vs Morph Cloud (2026)

An agent in a Daytona sandbox reaches outside systems through what runs inside it. Git operations clone and push repositories, secrets hold credentials and external storage can be mounted. Firewall allow lists decide which hosts the agent may reach. Daytona has no catalog of connectors, OAuth actions or tool gateway of its own. The tools are the customer's code.

Sourcedaytona.io/docs/llms.txtread 2026-09-22

An agent on Morph reaches outside systems through what runs in its VM. Secrets stored in the Morph account come preconfigured in devboxes, an agent token can let the agent read and write its own devbox's secrets, and volumes serve shared files to any S3 client. There are no connectors, tool catalog or MCP gateway for an agent to call.

Sourcecloud.morph.so/docs/devboxes/agent-tokensread 2026-11-05

The platform manages sandboxes. It creates and resizes them, applies lifecycle rules and runs warm pools and fleet scaling. It does not sequence, branch, retry or route an agent's steps. In Daytona's guides, agents built in LangGraph, the OpenAI Agents SDK and other frameworks do that, and Daytona has no workflow logic that mixes fixed steps with agent steps.

Sourcedaytona.io/docs/en/sandboxesread 2026-09-22

There is no workflow logic in Morph Cloud. Branching copies a running instance into many that continue from the same state, and TTLs stop or pause instances on a timer, but sequencing, retrying and routing an agent's steps stay in the customer's own code, which builds any fan out on these primitives.

Sourcecloud.morph.so/docs/concepts/mental-modelread 2026-11-05

Events go outward. Webhooks notify the customer's endpoint when sandboxes are created or change state and when snapshots and volumes change. Automated lifecycle rules stop, pause, archive or delete idle sandboxes. An agent's work starts only when the customer's code starts it. There is no schedule, and a stopped sandbox does not wake on an inbound request.

Sourcedaytona.io/docs/en/webhooksread 2026-09-22

A paused instance wakes without the customer's code calling it. With wake on HTTP, a request to an exposed service resumes the instance, and wake on SSH does the same for SSH sessions and exec calls, so an agent served from a Morph workspace starts when traffic arrives and pauses again when its TTL runs out. Morph's own guide builds agent workspaces this way. There is no scheduler or webhook.

Sourcecloud.morph.so/docs/documentation/instances/wake-onread 2026-11-05

The agent's code runs on Daytona, but there is no retrieval structure over the customer's content. Volumes and mounted external storage hold files the agent's code reads, and there is no index to query. The language server in a sandbox serves code intelligence for the files there. Nothing in Daytona indexes, refreshes or sets permissions on the customer's sources.

Sourcedaytona.io/docs/llms.txtread 2026-09-22

No retrieval index or knowledge base over the customer's content comes with the platform. Volumes and the VM filesystem hold files an agent's code reads, and any retrieval runs as the customer's own software inside the instance.

Sourcecloud.morph.so/docs/documentation/overviewread 2026-11-05

Persistence is on by default at three layers. The filesystem survives stop, start and archive. On VM sandboxes, memory (running processes and loaded state) survives pause and resume and hot snapshots. Volumes and mounted external storage outlive any one sandbox. Sandboxes can also be forked, and the agent's code reads the restored state to carry on. Paused process state belongs to the machine, so the customer can resume or delete it but cannot review or edit it. There is no separate memory layer that sets what an agent keeps or for how long.

Sourcedaytona.io/docs/en/persistenceread 2026-09-22

An agent's working state carries across sessions as machine state. A paused instance resumes with its memory intact, and a snapshot captures the whole running machine, processes included, so an instance booted from it carries on where the original stopped. Snapshot TTLs set how long saved states are kept, and volumes hold caches, checkpoints and datasets. This is machine state the developer manages, with no memory layer of stated scope that can be reviewed or deleted as memory.

Sourcecloud.morph.so/docs/developersread 2026-11-05

Work in a sandbox has no approval step, consent checkpoint or escalation to a person. VNC, a web terminal and SSH let a person watch a sandbox and step in by hand. A firewall on each sandbox (allow lists or block all, with organization tiers that sandbox settings cannot override) restricts what an agent's code can reach. Nothing holds an agent's action for a person to review.

Sourcedaytona.io/docs/en/network-limitsread 2026-09-22

Work in an instance has no approval step, consent checkpoint or escalation to a person. People can watch and step in by hand, since devboxes are shared live with the organization and open in VS Code, Cursor or a terminal over SSH, and a remote desktop guide serves a Linux desktop in the browser. Agent tokens limit what an agent can do to its own devbox but do not hold an action for review.

Sourcecloud.morph.so/docs/devboxes/sharing-and-collaborationread 2026-11-05

Organizations sign in through OpenID Connect providers such as Google Workspace, Microsoft Entra ID and Okta, with provisioning and membership handled at sign in on the Enterprise plan. Collaborative organizations have owner and member roles plus granular assignments by resource, such as SSO Admin and access to the audit log.

API keys carry permissions and scopes, with managed child keys. An audit log covers user and system activity, and each sandbox has firewall allow lists or block all. Daytona says it meets HIPAA, SOC 2 and GDPR standards and has a Trust Center, but it names no report type or auditor.

Sourcedaytona.io/docs/en/ssoread 2026-09-22

Organizations have admin and member roles with invitations, and one account can belong to several organizations. API keys act for the user, agent tokens confine an agent to one devbox and a set of scopes such as snapshots, HTTP services and secrets, and SSH public keys are registered with expiry dates, with an organization view of which keys have access. Morph's terms commit to encryption in transit and at rest, least privilege access and audit logging, and offer third party attestations such as SOC reports on request, with no certification named.

Sourcecloud.morph.so/web/legal/terms-of-serviceread 2026-11-05

Each step an agent takes in the sandbox is recorded. SDK tracing instruments every Daytona API operation and SDK call from the customer's application, such as each command run, file written or process started. Sandbox telemetry collects traces, logs and metrics from inside the sandbox.

Both export over OpenTelemetry to the customer's own collector, and Daytona keeps sandbox telemetry for three days. A separate audit log records user and system activity with the user, action and time, and logs can be streamed. The model's reasoning runs outside Daytona. Audit export is not available yet, and Daytona says it is coming soon.

Sourcedaytona.io/docs/en/observability/otel-collectionread 2026-09-22

The machine is measured and the agent is not. Each instance reports point in time CPU and memory use, usage and quota APIs break consumption down by service, command output streams back from exec calls, organizations can review which SSH keys have access, and the GitHub Actions console streams live events and logs for CI runs. Nothing traces or logs the steps an agent takes.

Sourcecloud.morph.so/docs/api-reference/get-instance-metrics-instance-instance-id-metrics-getread 2026-11-05

Sandboxes run in two shared regions, the United States and Europe, chosen per sandbox. Under Bring Your Own Compute, an organization instead creates custom regions and attaches its own runner nodes, with an optional proxy and snapshot manager for each region, for control over data locality and compliance.

Daytona's control plane manages them, and BYOC is sold on the Enterprise plan. The platform sold today cannot be self hosted from open source. Core development has moved to a private codebase, and the public github.com/daytonaio/daytona repository gets no further updates. Its last release, v0.190.0, stays public under AGPL 3.0.

Sourcedaytona.io/docs/en/bring-your-own-computeread 2026-09-22

Morph Cloud runs as a hosted service with no choice of region and no self hosted or customer cloud option, even though Morph's terms refer to region availability.

Sourcecloud.morph.so/docs/documentation/overviewread 2026-11-05

Ready made starting points for running agents include default sandbox snapshots in several sizes, a Declarative Builder for images and an agent skill for coding agents. More than forty guides run coding agents (Claude Code, Codex, Devin, Gemini CLI, OpenClaw, OpenCode), as well as framework agents and reinforcement learning setups, in sandboxes, and a reference implementation runs Cursor's Cloud Agents on self hosted Daytona sandboxes. Developers build from these environments and walkthroughs. There is no browsable catalog of agents a customer can adopt.

Sourcedaytona.io/docs/en/agent-skillsread 2026-10-23

Starting points come as templates and guides. Devbox templates bundle a base snapshot, build steps and integrations and can be shared with teammates by a short alias, and guides set up Claude Code workspaces, a browser based Linux desktop, Docker BuildKit, build caches and GitHub Actions runners. These are environments a developer builds from, not prebuilt agents a business adopts.

Sourcecloud.morph.so/docs/examples/claude-coderead 2026-11-05

The customer chooses the model. Guides run agents on Claude, OpenAI Codex and the Agents SDK, Gemini, Kimi and AWS Kiro, and on open models through OpenCode and other frameworks. GPU sandboxes serve the customer's own LLMs with vLLM or SGLang. Daytona imposes no model, and customers bring their own providers and keys, with guides on setting them up.

Sourcedaytona.io/docs/llms.txtread 2026-09-22

The customer chooses the agents and models that run on Morph. Any CLI agent can run in a devbox, Morph's guide installs Claude Code with the customer's own Anthropic key, and model calls happen in the customer's code. Morph supplies the machines, not the models.

Sourcecloud.morph.so/docs/examples/claude-coderead 2026-11-05

Outside callers reach Daytona through SDKs in five languages (Python, TypeScript, Ruby, Go and Java), a REST API with published OpenAPI specifications, and a CLI. The specifications cover the Daytona API, the Toolbox API inside the sandbox and an Analytics API. An MCP server lets Claude, Cursor and Windsurf manage sandboxes, and an agent skill serves coding agents.

Sourcedaytona.io/docs/llms.txtread 2026-09-22

Morph Cloud is driven through a REST API with a public OpenAPI specification, Python and TypeScript SDKs and the morphcloud CLI, which cover images, snapshots, instances, branching, devboxes, volumes, organizations, secrets, usage and SSH keys, and the CLI accepts plugins. Morph offers no MCP server.

Sourcecloud.morph.so/docs/api-referenceread 2026-11-05

Evaluations and reinforcement learning run on Daytona, but it does not score them. Its guides run training with HUD, TRL, veRL and OpenEnv in sandboxes, along with a benchmark agent across several languages. The datasets, reward functions and scoring belong to those frameworks. Daytona offers no fixtures, scoring or quality gates for a customer's agent workflows, and no way to compare agent runs.

Sourcedaytona.io/docs/llms.txtread 2026-09-22

There are no test cases, scoring or quality gates for an agent's behavior. Morph presents faster agent evaluations and capturing the exact state of a failed CI run as uses of snapshots and branching, with the datasets and scoring left to the customer's own harness.

Sourcecloud.morph.so/docs/developersread 2026-11-05

The Computer Use API gives an agent programmatic control of a desktop in the sandbox, from mouse and keyboard to screenshots, screen recording and display operations, on Linux and Windows. VNC lets a person watch or take over. The desktop runs remotely on Daytona's infrastructure. macOS sandboxes run on Mac hardware reserved for the customer through use.computer, with their own API key, and the Daytona SDKs, CLI and API do not manage them.

Sourcedaytona.io/docs/en/computer-useread 2026-10-23

No browser or desktop for an agent to drive comes with the platform. Morph's remote desktop guide installs an XFCE desktop with VNC and a noVNC web client so people can use it in the browser, and an agent that needs a browser installs and drives one itself inside the VM.

Sourcecloud.morph.so/docs/examples/remote-desktopread 2026-11-05