Comparison
SMELT Studio and Zapier
We barely compete. Zapier is bought by operations and marketing teams to connect an enormous catalogue of applications, and it is very good at that. This is a tool engineers buy when the hard part of the workflow is the decision in the middle. If the problem is reaching the applications your company already runs on, buy Zapier and do not think about it again.
Different buyer, different job
Zapier connects a catalogue we are not going to match, carries an enterprise brand and compliance posture we do not have, and is very probably already on your company card. None of that is contested here and none of it should be.
The question worth asking is what happens in the middle. Zapier is built for trigger to action: something happened, do this. It is not built for a workflow where the middle step has to weigh an ambiguous input, choose between several courses of action, and be inspectable afterwards when someone asks why it chose that one.
We are not asking anyone to cancel Zapier. Most sensible stacks keep it.
Code steps
Zapier includes Code steps in Python and JavaScript with a runtime allowance per step. Beyond that allowance, their published pricing says the extra time is
rounded up to the next 30-second block, with each block using 1 task.
The SMELT Code block is a deliberately different shape. It runs JavaScript compiled to WebAssembly inside a wasmtime sandbox with no network access, no filesystem, and no ambient capabilities at all. Processor time is bounded by a fuel budget, wall clock by an epoch deadline, and memory by a fixed ceiling, and the clock the script sees is frozen so it cannot time itself out of the sandbox. Output size is capped. It is validated in the editor before you run it.
The point is not that one is more generous. It is that ours is a bounded execution environment you can put in the middle of an agent workflow without wondering what it can reach.
A canvas you can read afterwards
Zapier's agent product is a chat-driven assistant over its own application graph, billed separately from tasks in what their pricing page calls activities:
any billable action an agent takes: using a trigger, answering from a knowledge source, running an action, web browsing or scraping, a web search, or a message sent via the Chrome extension.
For simple tasks that is easier than building anything, and we would not pretend otherwise. Zapier Agents do keep a record of what an agent did: an activity view is one of the features they list.
The difference is what the record is checked against. Theirs describes what an assistant chose to do. A SMELT workflow is a graph you laid out first, so the trace of every block, its output, its duration and its credits reads against a structure you authored, alongside the retries and error paths you configured and an approval gate you can put in front of anything irreversible. When something surprising happens, "why did it do that" has an answer you can point at on the canvas rather than a transcript to interpret.
What a run costs, while it is running
Runs are metered in credits. Each block reports its credits and duration as it finishes, the trace shows the credits a delegated agent call consumed, and your workspace usage shows credits drawn against the monthly allowance on your plan. The number is in the app while the work happens rather than on a model provider invoice four weeks later.
This is not a claim about being inexpensive, and we would rather you did not read it as one. It is one predictable number covering orchestration, inference, retries and key management. If your objection is price per token, connect your own OpenAI or Anthropic key and the meter stops entirely.
Plans are monthly. See pricing for the allowances on each tier.
The answer to "do you support X"
If X has a REST API, you can reach it today. The HTTP block calls any endpoint, with your own headers, query parameters and JSON body, and the response comes back as a typed output the next block can reference.
It is guarded rather than open. The host is resolved once, private, loopback and link-local addresses are refused, and the request is then sent to the address that was checked, with the original host header preserved. A DNS answer that changes underneath you does not turn into an SSRF against our network or yours.
Inbound, a webhook trigger gives a workflow its own ingest URL and a signing secret you send in the x-webhook-secret header, and a Respond to Webhook block lets the workflow answer the caller in the same request. Schedule, form, chat and manual triggers sit alongside it, as do triggers for the providers in the palette.
This is a capability, not an apology for a short list. It is also why we do not publish a connector count: it would be a scoreboard for a game we deliberately are not playing.
Who should not buy this
If your team is not technical, use Zapier. If the value is the catalogue, use Zapier. If nobody on your side will ever open a trace, most of what we have built is wasted on you and you will be paying for it.
What we do not have
Better you read this from us than find it in a comment thread. Each of these ends some evaluations, and it should.
- No self-host. There is no way to run SMELT Studio on your own infrastructure. Our typed contracts are published and inspectable, so the shape of the architecture is not a secret, but that is weaker than being able to run the thing yourself and we are not going to dress it up as an advantage.
- No Data Tables. There is no built-in store for structured rows between runs. Keep that state in a database you already own and reach it over HTTP.
- No evaluations surface. There is no place in the app to assemble a test set for an agent, score it, and watch the score move when you change a prompt. If a graded regression suite is a condition of putting an agent near a customer, we do not meet it yet.
- No embeddable chat widget. You cannot drop a SMELT chat onto your own website. A hosted form is the only public entry point we offer, and it is a form, not a conversation.
- No MCP server. SMELT can call an external MCP server as a tool. It cannot be one, so another AI client cannot reach into SMELT and run your workflow that way.
- Agent memory is a capped list. An agent decides what to remember, and recall is by semantic search rather than a raw read-back. The list is still capped per agent, and the least recently used entries fall off the end. Treat it as recall between runs, not as a knowledge store.
If one of these is a condition of your buying, say so at [email protected] rather than working around it.
Coming from a general automation tool
The concept mapping in the n8n migration guide is written for n8n, but the vocabulary of triggers, steps, branches and credentials carries over well enough to be worth reading first.
Sources
Every claim about another product on this page is checked against that product's own published pricing or documentation, and the date of the last check is recorded in this page's source. Competitors ship quickly, and a page like this goes stale by default rather than by neglect. If something here has gone out of date, tell us at [email protected] and we will correct the page.
Try it against your own workflow
The free tier runs real workflows. Build the one you are unhappy with elsewhere and see whether the second change is easier than the first.