AI-native chat platform on Cloudflare
Ship AI-powered chatacross 14 platforms.
One SDK for in-app chat, AI agents, and multi-platform messaging. Streaming markdown, MCP tool calling, and LLM middleware — self-host on Cloudflare Workers or use hosted cloud.
pnpm add @fluxy-chat/sdkYour app
Live preview
The same hooks power your production UI: wire the client once, then iterate on layout and copy — the preview mirrors the room your users join in your app.
Works with stacks you already use
One SDK across common frontends and runtimes. Fork the repo when you want full control.
Featured on
Deploy in your account. Quotas live in D1. Console access can require a one-time ack — usage still runs through your Worker with your keys.
14 platforms
Slack, Discord, Telegram, WhatsApp, Teams, and 9 more — one unified adapter interface.
AI-native
Streaming markdown, tool calling, MCP client, and LLM middleware pipeline built in.
30+ SDK modules
Rooms, presence, typing, history, agents, webhooks, cards, and more — all on @fluxy-chat/sdk.
Durable agents
WorkflowAgent persists to D1, survives deploys, and resumes from the last step.
Realtime voice
Bidirectional voice-to-voice AI with provider-agnostic abstraction.
MIT self-host
Deploy the Worker + D1 in your Cloudflare account. Read every line of source.
Messaging basics, on the edge
Channels, presence, mentions, and webhooks without bolting on another realtime vendor. Tap or hover a row on desktop.
Threads, polls, and translation are up to your product layer. FluxyChat ships the realtime core — rooms, presence, typing, delivery, and AI agent events.
What ships in the box
Edge delivery, AI-native streaming, 14 platform adapters, and a console for day-two work. You own the UI; we handle rooms, delivery, agents, and ops hooks.
Most products stitch together chat, AI, support, and governance from different vendors. That stack gets expensive, hard to audit, and painful to migrate. FluxyChat combines 14 platform adapters, AI-native streaming, MCP tool calling, LLM middleware, and compliance hooks on the edge you already run.
- 14 platform adapters (Slack, Discord, Telegram, WhatsApp, Teams, +9)
- AI-native streaming with tool calling & HITL approval
- MCP client for external tool servers
- LLM middleware pipeline (guardrails, caching, RAG, PII redaction)
- WorkflowAgent for durable agent execution
- Card builder with Slack Block Kit & Teams Adaptive Cards
- DevTools playground & OpenTelemetry tracing
- GDPR export, audit trails, and retention policies
Already set up? Open the console for rooms, agents, and billing.
Built-in architecture
14 platform adapters
Slack, Discord, Telegram, WhatsApp, Teams, Email, SMS, Webhook, Matrix, and 5 more — one unified interface
Streaming markdown
Table buffering, code fence tracking, inline marker healing — clean partial renders during AI streaming
MCP client + tool calling
Consume any MCP-compatible tool server. Auto-convert to LLM function-calling format with HITL approval gates
Card builder
Composable rich messaging with JSX or function API — Slack Block Kit and Teams Adaptive Card renderers built in
Middleware pipeline
Transform messages and AI responses before they land
Moderate, validate, enrich, and fan out from the edge. LLM middleware supports guardrails, caching, RAG injection, PII redaction, and logging via wrapGenerate / wrapStream / transformParams. Policy code stays on the data path, not in a sidecar you forget to deploy.
Developers
SDK, CLI, docs, and console in one repo
Auth, rooms, and retries sit behind hooks and route handlers. Scaffold a new project with create-fluxy-chat CLI. When you need keys, quotas, or billing, open the operator UI instead of wiring another admin surface.
Where teams start
AI-native architecture overhaul
AI-native architecture inspired by the Vercel Chat SDK and AI SDK. Cleaner abstractions, richer streaming, MCP tool calling, LLM middleware, and durable agent execution — without losing the real-time, multi-tenant, enterprise depth FluxyChat is known for.
Enterprise-ready
Self-serve plans cover most production traffic. When governance matters, these capabilities are built into the worker — not a separate SKU bolted on later.
- SSO / SAML & SCIM provisioning
- Audit logs with export schedules
- Retention policies & legal hold
- IP whitelisting & DLP integrations
- GDPR export and erasure tools
- Signed webhooks with retry & delivery logs
- Per-tool approval gates for AI agents
- OpenTelemetry tracing with GenAI semantic conventions
Explore capabilities in the features console.
Pricing
Self-serve tiers with monthly message, agent, and webhook quotas. Platform fee today; usage add-ons on Growth and above when you outgrow included limits.
Console routes can require a one-time ack on your dashboard host. Billable usage still needs your Worker credentials.
Self-serve — checkout today
Free includes 200,000 messages, 5,000 agent invokes, and 50,000 webhook deliveries /mo. Pro raises each to 10× for $50/mo — ideal for production rooms with active agents.
Free
Try the stack without a card.
- 200,000 messages / month
- 5,000 agent invokes / month
- 50,000 webhook deliveries / month
- SDK and dashboard access
- Community support (GitHub + Discord: https://discord.gg/fluxychat)
- One project, fair-use limits
- Upgrade without migrating data
Starter
Production traffic with fixed monthly limits.
- 500,000 messages / month
- 10,000 agent invokes / month
- 100,000 webhook deliveries / month
- Higher quotas than Free
- Signed webhooks with retries
- GDPR export and erasure endpoints
- Email support (best effort)
Pro
Heavy rooms, agents, and webhook volume.
- 5,000,000 messages / month
- 100,000 agent invokes / month
- 1,000,000 webhook deliveries / month
- 5M messages, 100k agent invokes, 1M webhook deliveries per month
- Priority support
- Custom retention and audit (contact us)
- Annual invoicing on request
Team
Multiple projects, more seats, same compliance posture as Pro.
- 20,000,000 messages / month
- 200,000 agent invokes / month
- 1,000,000 webhook deliveries / month
- 20M messages, 200k agent invokes, 1M webhook deliveries per month
- Up to 5 projects and 5 team seats
- Team member management and RBAC
- Priority support, annual invoicing on request
Growth
B2B SaaS with omnichannel, AI memory, and FTS search.
- 100,000,000 messages / month
- 1,000,000 agent invokes / month
- 5,000,000 webhook deliveries / month
- Multiple projects, omnichannel inbox (base)
- AI memory and full-text search
- Light moderation tooling
- Annual invoicing on request
Sales-led — governance and higher limits
Business
Serious teams needing SSO and audit at scale.
- SSO add-on
- Custom domain and higher limits
- Audit export schedules
- Priority support, white-label lite
Enterprise
Regulated and large deployments.
- SCIM and SLA
- Retention and DLP pack
- Compliance review
- Dedicated support
Why usage-based quotas?▾
Real cost scales with messages, AI invokes, and webhook volume. Self-serve plans include fixed monthly quotas; heavy usage can be metered on Growth and above.
Do AI agent invokes count against my quota?▾
Yes — each plan includes a monthly agent invoke limit. AI streaming, tool calls, and MCP interactions all count as invokes. Pro raises the limit 10× for production copilots and agentic features.
Enterprise vs Business?▾
Business is high-limit self-serve with SSO add-on and audit export. Enterprise adds SCIM, SLA, DLP, security review, and dedicated support.
Self-host vs hosted?▾
Same worker and SDK. Hosted is fastest to start; self-host gives full control on your Cloudflare account with MIT source.
Need enterprise SSO, VPC-style isolation, or custom SLOs? Email fluxychat@outlook.com.
How we compare
Not a feature shootout — rough fit vs Stream, Ably, Pusher, and DIY DO repos. Full compare · Leaving Pusher.
| Capability | Typical stream APIs | Typical Ably-style | Typical Channels (Pusher-style) | Fluxychat |
|---|---|---|---|---|
| Edge-native (Cloudflare Workers + DO + D1) | Managed cloud | Managed cloud | Managed cloud | Designed for Workers + DO + D1 |
| Multi-platform adapters (14 platforms) | Limited / separate product | N/A | N/A | Slack, Discord, Telegram, WhatsApp, Teams, +9 — unified interface |
| AI-native streaming (markdown, tool calling) | Add-on / separate product | N/A | N/A | Streaming markdown, tool_call/tool_result, HITL approval |
| MCP client integration | N/A | N/A | N/A | Consume any MCP tool server — auto-convert to function-calling |
| LLM middleware pipeline | N/A | N/A | N/A | wrapGenerate / wrapStream / transformParams — guardrails, RAG, PII |
| Durable agent execution (WorkflowAgent) | N/A | N/A | N/A | Persists to D1, survives deploys, auto-resume from last step |
| In-app chat + operator console | Separate product areas | Console + APIs | Channels dashboard | First-party console in monorepo |
| Headless SDK (optimistic sends, reconnect state) | Strong SDKs | Strong SDKs | Channels SDKs | @fluxy-chat/sdk + vanilla store |
| Agent tool events on room WebSocket | Varies | Separate products | N/A | tool_call / tool_result on same timeline |
| Message templates + member preferences API | Varies | N/A | Limited | POST /templates, member prefs PATCH |
| Reconnect, replay, and delivery state in SDK | SDK features vary | SDK features vary | Channels SDK | connectionState, loadMore, clientMessageId idempotency |
| Read receipts / unread badges | Product features vary | Varies by product | Not first-class in Channels | markReadLatest + room list unread in SDK/console |
| In-app notifications (mentions, DMs) | Separate notification products | Separate products | Beams (separate SKU) | REST + SDK useNotifications on same Worker |
| Message middleware (validate / filter / enrich) | Varies | N/A | Webhooks only | Edge pipeline before persist + broadcast |
| Pricing surprises at scale | Enterprise / usage tiers | Usage-based | Free tier small; connections add up | Cloudflare pricing you can read; MIT self-host option |
| Self-host / on your own account | Proprietary cloud | Managed-first | Managed-first | Full MIT monorepo — deploy Worker + D1 in your CF account |
| Socket fleet / VPS to operate | Managed vendor infra | Managed vendor infra | Managed vendor infra | No VPS; one Room DO per room on CF edge |
| Next.js on Vercel + realtime (typical split) | Managed cloud + your frontend | Ably + Vercel tutorial pattern | Channels + serverless functions | Vercel/Netlify UI + CF Worker chat (no Vercel WS limits) |
What does it actually cost on Cloudflare?
Real numbers, not vague "contact sales" ranges. The free tier is generous; the paid tier scales with usage; self-host is just your Cloudflare bill. No per-connection pricing surprises.
FluxyChat runs entirely on your Cloudflare account. You pay Cloudflare directly no per-seat or per-message markup from us on the self-host path.
| Plan | Monthly messages | D1 reads / day | Cost (USD) | Best for |
|---|---|---|---|---|
| Workers Free | ~3M / mo (100k req/day × 30) | 5M / day | $0 | Hobby · early beta · <1k MAU |
| Workers Paid ($5/mo) | ~300M+ / mo | 25M+ / day | $5 + usage | Growing SaaS · 1k–100k MAU |
| Self-host on your CF account | Unlimited* | Your plan limits | Your plan cost | Enterprise · cost-sensitive · compliance |
* D1 storage cap: 10 GB on the free plan, unlimited on paid. Message fanout uses Durable Object egress see Cloudflare's current pricing for live numbers. Run pnpm run check:pricing to verify your env assumptions match the published rates.
FAQs
Questions we hear when teams compare hosted chat APIs and edge deployments.
What is the chat API vs the SDK?▾
Your backend talks to the Worker over HTTP. The browser uses @fluxy-chat/sdk to join a room, send, and render. Same product, two surfaces.
Can I keep the frontend on Vercel?▾
Yes. Lots of teams leave Next on Vercel or Netlify and run chat on Workers + Durable Objects. You skip the platform WebSocket limits and a second Pusher-style bill.
Is this Intercom, Zendesk, or a live-chat widget?▾
No. Those are support desks. FluxyChat is the room layer for chat inside your app — SDK, webhooks, agent timeline. You build the UI. Guide: /guides/in-app-chat-vs-support-desk.
Is this a Pusher or Ably replacement?▾
For tenant-scoped in-app chat on the edge, usually yes: rooms, history, reconnect, operator tools. For generic pub/sub or SMS/WhatsApp, keep those products and use FluxyChat for the chat slice.
How is this different from a fully managed vendor?▾
You can run the Worker and D1 in your Cloudflare account, read the MIT source, and control upgrades. Hosted cloud is there when you do not want to wire infra on day one.
Can I add moderation and webhooks?▾
Yes. Middleware on the edge, signed webhooks, and console pages for the boring admin work.
What are Durable Objects for chat?▾
Cloudflare puts shared state in a DO — chat rooms are the textbook case. FluxyChat uses one Room object per room and D1 for history. Guide: /guides/durable-objects-for-chat-rooms.
Humans and AI agents in the same room?▾
tool_call and tool_result ride the same WebSocket as user messages, so you replay one timeline instead of guessing what the agent did. Streaming markdown with table buffering and code fence tracking renders cleanly during AI responses. Guide: /guides/agent-events-same-websocket-stream.
What platforms do the multi-platform adapters support?▾
14 platforms behind a unified interface: Slack, Discord, Telegram, WhatsApp, Microsoft Teams, Email, SMS, Webhook, Matrix, and more. Each adapter handles platform-specific formatting, and the card builder renders Slack Block Kit and Teams Adaptive Cards natively.
What is MCP and how does FluxyChat use it?▾
Model Context Protocol lets your agents consume external tool servers. FluxyChat acts as an MCP client — auto-converting tools to LLM function-calling format. Point your agent at any MCP-compatible server and it just works.
Can I add guardrails or RAG to AI responses?▾
Yes. The LLM middleware pipeline supports wrapGenerate, wrapStream, and transformParams hooks. Plug in guardrails, caching, RAG injection, PII redaction, or logging — all on the edge, before responses reach users.
Does WorkflowAgent survive deploys?▾
Yes. WorkflowAgent persists state to D1 after each step. If the Worker restarts or you deploy a new version, the agent resumes from the last completed step — no lost work.
I finished Cloudflare's chat tutorial — now what?▾
That tutorial is the right start. FluxyChat adds JWT, D1 history, reconnect in the SDK, and a console — /guides/after-cloudflare-chat-tutorial.
Why not stay on Pusher?▾
You might not need to. Edge pricing you can read, self-host on your account, no socket VPS fleet. Compare at /compare.
Where should I start?▾
Run the quickstart for a JWT and first room, or read /guides if you are still picking between DIY Workers chat and a packaged layer.
Does a public dashboard mean anyone can spend my quota?▾
No. Billable calls need your JWTs and API keys. Turn on DASHBOARD_ACCESS_MODE=ack (and optional CONSOLE_GATE_SECRET) so console routes need a one-time ack first.
Build AI-native chat on infrastructure you own.
Start on the free tier with 14 platform adapters, streaming AI, and MCP tools. Run a pilot on hosted cloud, or deploy the worker in your Cloudflare account when you are ready.