Skip to content
RSFlowHub
Developer docs API v1 Stable OpenAPI SDKs Playground
Developer Documentation & Unified AI API Reference
Developer Documentation

Build with RSFlowHub AI

Integrate powerful AI capabilities into your product using our REST APIs. Everything you need - authentication, live examples, and reference docs.

Platform overview

How RSFlowHub works

Production-grade AI gateway designed for deterministic machine-to-machine integrations.

Automatic Provider Fallback

Direct System Definition

RSFlowHub is a developer-first REST microservice platform that standardizes access to modern large language models. Rather than writing prompts and parsing unformatted text completions, developers call specialized endpoints (such as intent detection, sentiment analysis, email workflow generation, and translation) that return clean, structured JSON envelopes. Requests usually complete in 1 to 3 seconds, fail over to a backup AI provider automatically, and are billed through atomic credit pre-authorization so you are never charged for failed calls.

Architecture
REST over HTTPS
Authentication
x-api-key (SHA-256)
Target Latency
1-3 s typical
Default Rate Limit
60 req / min

Core Production Integration Rules

60-Second Timeout

Never let HTTP clients wait forever, but allow 60 seconds: a call can wait for the AI and then try a backup model. The official SDKs use this default.

Exponential Jitter Backoff

Retry 429 and transient 5xx errors with randomized jitter to eliminate thundering herd storms.

Pre-Auth Credit Locks

Credits are verified upfront and decremented atomically upon completion. Zero billing on validation failures.

Start here

Official SDKs

Call every API from PHP, Node.js / TypeScript or Python with one package. The SDKs handle authentication, typed errors, retries and file uploads, and new APIs work without updating them.

terminal
composer require rsflowhub/rsflowhub-php
npm install rsflowhub
pip install rsflowhub

AI Chatbot Widget

Add an AI chatbot trained on your own content to any website with one script tag; no code to write. Colors, greeting, quick replies, lead capture and support tickets are all set in the dashboard, and changes appear without touching your site again.

before </body> HTML
<script src="https://rsflowhub.com/widget/v1/rsflowhub-widget.js"
        data-widget-token="wt_sec_YOUR_WIDGET_TOKEN"
        async></script>

Contact Forms & Newsletter

Drop a spam-protected contact form or newsletter signup into any page with one script tag, or post to the Forms API from your own form. Messages and subscribers land in your dashboard, with email alerts, AI spam checks and webhooks.

where the form should appear HTML
<script src="https://rsflowhub.com/widget/v1/rsflowhub-forms.js"
        data-form-token="fm_sec_YOUR_FORM_TOKEN"
        async></script>

All AI APIs

Every endpoint takes JSON and returns the same {success, request_id, data, meta} envelope. Use the search box above to filter.

Frequently Asked Technical Questions

RSFlowHub is a platform of ready-made AI APIs. Each endpoint (intent detection, sentiment analysis, translation, moderation, email and ticket automation, data extraction and more) takes a JSON request and returns a structured JSON result, running on models from OpenAI, Google Gemini, DeepSeek and xAI Grok with automatic fallback. Usage is billed in credits, with a free plan to start.

All requests require header authentication using the x-api-key HTTP header (or Authorization: Bearer <API_KEY>). Keys are hashed with SHA-256 for maximum security and can be rotated instantly via the developer dashboard.

RSFlowHub prevents unexpected overdrafts and billing race conditions using atomic pre-authorization locks. Before executing your request, the gateway verifies that your account has the base credits required. If the request is rejected during validation (422) or is malformed (400), no credits are charged.

Most requests complete in 1 to 3 seconds depending on input length and model; identical repeats on cache-enabled endpoints return faster. A circuit breaker retries failed provider calls on a backup model. Use a 15 to 30 second client timeout and exponential backoff with jitter for retries.

Ready to build?

Create your free account and make your first API call in minutes.

RSFlowHub

Sign in to continue

You'll come straight back to this page.

Forgot password?

New to RSFlowHub? Create a free account