n8n is the incumbent of self-hosted automation — mature, broadly integrated, and its “see and control” positioning is real at build time. The difference shows up after you press run.
An n8n execution runs the workflow as persisted in the database — unsaved canvas state is ignored — and on production (non-manual) executions the canvas doesn't even receive status updates. LiveGraph's engine re-reads the live graph on every hop, so the canvas you're watching is the thing being executed.
| Dimension | n8n | LiveGraph |
|---|---|---|
| What a run executes | The persisted workflow snapshot — edits made on the canvas since save are ignored by the running execution. | The live graph, re-read on every hop — what you see is what's executing. |
| Canvas during a production run | Inert — non-manual executions don't stream status to the canvas (per n8n's own engineers). | Stays live — every hop is visible while it executes, and dragging an edge changes where this run goes next. |
| Mid-run rerouting | Not supported — a wrong route means a failed or cancelled run. | Drag an edge; the next hop follows the new route. |
| Cancelling a run | In queue mode, cancellations don't propagate to the workers executing the run (open issue). | A cancel reaches the executing hop. |
| Human approval gates | Not evaluated | Run-level gates that park at the approvals queue, plus per-tool-call gates (e.g. a single http_request endpoint) — the pending payload is always visible to the approver. |
| Model roster | Not evaluated | Per-node provider and model — Anthropic, OpenAI, Google, Groq, OpenRouter, or an OpenAI-compatible endpoint; your call per node. |
| Run audit trail | Not evaluated | Every hop lands as an ordered run event — input, output, the routing explanation, and its cost — and approval decisions are stamped with who and when. |
| Pricing | Fair-code (source-available) license; self-hosting is a first-class, widely-run option. | Flat team price — runs are never metered. Model usage on hosted keys is metered on top; your own keys are never marked up. |
Competitor rows are limited to claims verified in our own research — "Not evaluated" means we didn't check, not that the capability is missing. On the LiveGraph side, one nuance: runs started from chat or the canvas Run button are steerable; unattended runs (API, schedules, webhooks) run pinned to the graph they started on — steering is a watched-run feature, by design.
If your workflows are agent teams — models routing to models — the question isn't which canvas is nicer to draw on, it's whether the canvas is still alive when the run needs you. That's the thing LiveGraph does that n8n's engine doesn't.
The live demo runs a real multi-agent graph in your browser — no signup, no API key — including the mid-run edge drag. If the canvas earns it, an account takes a minute.