Skip to main content

workload / Checked 2026-10-18

Notte vs Steel for long browser workflows

A browser stays open while a workflow navigates through many pages or waits for a long operation. Compare Notte and Steel on the mechanism this job needs, the work that would move and a specific acceptance case.

The workflow decision

For long browser workflows, the useful comparison is whether Notte or Steel can meet the following outcome: The original job reaches its expected terminal state within the selected account's browser-duration entitlement.

Per-action latency, session duration and reconnect behavior matter more than a headline startup number. An idle timeout and a maximum lifetime can end the run for different reasons.

A proposed acceptance case

A job visits several pages, starts a permitted long operation and waits for a result. Progress must survive ordinary pauses without silently creating a second job.

This is an evaluation design, not a claim that this workflow has been executed successfully on every provider.

Part of the testRequired observation
Controlled inputA controlled multi-page workflow, a maximum wall-clock budget and an application job identifier.
Pass conditionThe original job reaches its expected terminal state within the selected account's browser-duration entitlement.
Negative caseDelay one step beyond the caller's wait budget. Inspect whether the browser continues, stops or needs explicit cancellation; do not submit the operation twice.
Evidence to retainStep checkpoints, original application job identifier, browser lifetime and final output.

Documented mechanisms, side by side

These labels describe product mechanisms. The linked source and the acceptance case determine whether a mechanism meets this workflow's requirement.

RequirementNotteSteel
Browser controlDocumentedDocumented
Saved browser stateDocumentedDocumented
Live viewDocumentedDocumented
Deployment choiceEnterprise optionsDocumented

Sources: docs.notte.cc [1] ↗docs.notte.cc [2] ↗docs.notte.cc [3] ↗docs.notte.cc [4] ↗docs.steel.dev [5] ↗docs.steel.dev [6] ↗docs.steel.dev [7] ↗docs.steel.dev [8] ↗

Browser control

Notte: Create an owned browser session, connect Playwright or Puppeteer over CDP, and stop the session when the work finishes. Browser lifecycle management remains necessary even when the control socket disconnects.

Steel: Steel sessions expose browser connection details for existing automation code. Create and release the owned session through Steel's lifecycle API; a browser library is the controller, not the infrastructure scheduler.

Sources: docs.notte.cc [1] ↗docs.steel.dev [2] ↗

Deployment ownership

Notte: Use a remote browser from your existing process or deploy the browser automation as a Notte Function. These choices put lifecycle, scheduling and monitoring responsibilities in different places.

Steel: Compare managed Steel with the documented self-hosted path explicitly. Self-hosting gives the team infrastructure responsibility that a hosted browser subscription would otherwise cover.

Sources: docs.notte.cc [1] ↗docs.steel.dev [2] ↗

Pricing model

Notte: The public usage rates are $0.01 per browser-hour, $0.05 per Function-hour and $10 per GB of managed proxy traffic. Paid plans return their price as usage credits. Price the relevant meters and plan eligibility for the workload.

Steel: Hosted browser plans and a self-hosted browser have different cost structures. For self-hosting, include the worker, compute, networking, storage and operations rather than assigning the open-source code a zero operating cost.

Sources: notte.cc [1] ↗docs.steel.dev [2] ↗

What should change the choice?

Evaluate Notte when session ownership needs to sit alongside Python execution or authenticated state. Check the eligible session duration before a low hourly rate influences the choice.

Steel: Steel's managed and self-hosted choices make infrastructure ownership a concrete deciding factor. Keep that option on the shortlist when operating the browser is intentional; evaluate Notte when owning fewer execution and authentication components is the actual goal.

A provider with a proven long-running session contract may be the better choice when moving would change timeout or continuation behavior. Startup rankings have less weight once a browser is already running.

Sources: docs.notte.cc [1] ↗docs.notte.cc [2] ↗notte.cc [3] ↗docs.steel.dev [4] ↗docs.steel.dev [5] ↗docs.steel.dev [6] ↗

The cost boundary

Use the same attempted work and eligible account capacity in both estimates. The browser-hour rate does not include every proxy, execution or identity meter, and the cost of an incomplete job is still part of the workflow budget.

The workload evidence should include: Step checkpoints, original application job identifier, browser lifetime and final output.

Sources: notte.cc [1] ↗docs.steel.dev [2] ↗

Make a bounded pilot

Run the full workflow for its expected duration and verify progress, reconnect behavior and final completion.

Use the same authorized account, input and final-state check in each configuration. Treat a missing product contract as an open question; do not turn an undocumented capability into an unsupported claim or a successful test result.

  • Record the source workflow's longest wait and the owner of each timeout.
  • Move one long job without changing its application identifier or completion check.
  • Keep the source runner available until target cancellation and final cleanup are verified.

Keep comparing

Put the workflow to the test.

Try Notte Read the quickstart ↗