AI & MLReal-timeUpdated May 2026

Replicate

Replicate open-source model inference

Replicate hosts open-source ML models behind a clean REST API — every model on the platform (Flux, SDXL, Whisper, Llama variants, niche specialist models for face restoration, voice cloning, video upscaling) gets the same Run-then-Poll interface. Tiny Command exposes four actions, no triggers: Run Prediction (start an inference against any model version, returns a prediction ID immediately), Get Prediction (poll for status — starting, processing, succeeded, failed, canceled — and retrieve the output URL(s) when done), List Models (browse the Replicate catalog), Cancel Prediction (abort an in-flight run). The connection uses a Replicate API token from replicate.com/account/api-tokens. Predictions are async by design — most generation models take 5 seconds to several minutes depending on model and input. The standard workflow: Run Prediction → poll Get Prediction every few seconds (or use Replicate's webhook by passing webhook_url) → download the result URL. Replicate's per-second billing matches actual GPU usage, so cheap-but-slow models can be cheaper per-output than fast-but-expensive ones.

Connect Replicate

0triggers

4actions

≈ 2 minto set up

Freetier · no card

Actions

Do anything Replicate can do, from a workflow.

Every action accepts dynamic inputs from upstream nodes, whether that's an AI output, a form field, or a search result.

Action	What it does	Inputs
Cancel Replicate Prediction	Aborts an in-progress prediction. For "user cancelled the request → stop the long-running model" workflows that save GPU compute costs.	1 fields	View →
Get Replicate Prediction	Polls a prediction by ID; returns status (starting/processing/succeeded/failed) and output URLs when complete. Configure webhook on Run Prediction for push notification.	1 fields	View →
List Replicate Models	Returns the Replicate catalog. Useful for model discovery — Replicate hosts many niche specialised models (face restoration, voice cloning, etc.) alongside the common foundation models.	1 fields	View →
Run Replicate Prediction	Submits an inference request against any model on Replicate. Returns prediction ID immediately; poll Get Prediction or use webhook for completion notification.	3 fields	View →

Recipes

Pre-built Replicate workflows.

Clone any recipe and customize it in one click. Every recipe is fully editable.

Replicate → Slack alerts

AI-triage every Replicate event, ping the right channel only when it matters.

Replicate → Google Sheets log

Every event matching a filter, appended to a running spreadsheet.

Replicate → Notion database

Turn Replicate into a Notion-backed source of truth, auto-tagged.

Before you build

Three things worth knowing.

Filter at the trigger

Tiny Command counts a run the moment a trigger fires. Filtering early means only matching events spend your usage budget.

Authorize once, reuse anywhere

Connect Replicate once and every workflow on your account can use its triggers and actions. You don't have to re-auth per workflow.

No JSON to read

Every Replicate field shows up in the visual picker for downstream nodes. The raw payload is there for power users, optional for everyone else.

FAQ

Questions about the Replicate integration.

If we missed yours, ping support. We usually reply within an hour.

How do I connect Replicate to Tiny Command?

Open the Tiny Command workflow builder, drop in a Replicate node, and click Connect. Authorize Replicate once and any workflow on your account can use its triggers and actions. Most teams finish the connection in under two minutes.

What Replicate triggers does Tiny Command support?

Tiny Command focuses on outbound actions for Replicate today. Use Tiny Command's universal Webhook or Schedule trigger to start Replicate workflows, then run any Replicate action you need.

What Replicate actions can I run from a workflow?

4 Replicate actions are available out of the box, covering ai & ml operations like "Cancel Replicate Prediction". Every action accepts dynamic inputs from upstream nodes, whether that's a search result, an AI output, or a form field.

Is the Replicate integration real-time?

Replicate actions execute synchronously inside your workflow. Tiny Command waits for Replicate's API to confirm before continuing to the next step, so downstream nodes can rely on the result.

Do I need to write code to use Replicate with Tiny Command?

No. Every Replicate trigger and action is fully configurable from the visual workflow builder. For edge cases that aren't covered, drop in a custom HTTP node and call any Replicate API endpoint directly.

How much does the Replicate integration cost?

There's a free tier you can start on without a credit card. Higher run volumes and team features come with paid plans. The Replicate integration itself has no per-app surcharge.