Let's make it real.
Start free, no credit card. $5 in credits, then $5 per 1,000 executions, with a spend cap you control. Unlimited Surfaces (deployed channels) on every plan. Scale to the Team plan as you grow.
No credit card. Start with $5 in free credits, then pay only for what you run.
or $3,999/year — 2 months free
For teams shipping real volume. $399/mo includes 300,000 executions at a much lower rate than Pay as you go, plus BYOK and support.
For platforms, agencies, and enterprises embedding Runtype at scale — your cloud, your keys, white-labeled and multi-tenant.
Tap tabs above to compare plans
No credit card. Start with $5 in free credits, then pay only for what you run.
or $3,999/year — 2 months free
For teams shipping real volume. $399/mo includes 300,000 executions at a much lower rate than Pay as you go, plus BYOK and support.
For platforms, agencies, and enterprises embedding Runtype at scale — your cloud, your keys, white-labeled and multi-tenant.
See what you'd pay
Set your expected monthly usage. Pay-as-you-go scales with what you run; Team becomes the cheaper choice once you grow.
Pay as you go
50k executions × $5/1k
Team
Estimates only. Managed model usage is billed separately (provider price + 20% platform fee); bring your own keys on Team to pay providers directly.
Compare all options
| Feature | Pay as you goStart here $0 + usage | TeamBest value over ~80k/mo $399/mo | EnterpriseCustom |
|---|---|---|---|
Unlimited | Unlimited | Unlimited | |
Metered ($5/1k) | 300,000/mo included | Custom volume | |
$5/1k | $3/1k after pool | Custom | |
Unlimited | Unlimited | Unlimited | |
100 | 200 | Unlimited | |
Daily | Per-minute | Unlimited | |
1 day | 90 days | Custom (1 year+) | |
$5 (one-time) | $50/mo (bank up to $100) | Custom | |
No surprise bills
Every plan includes unlimited Surfaces — you're billed by usage, not by how many channels you deploy. Set a monthly spend cap and get alerts as you approach it, so costs stay predictable.
AI product experiences like the ones shipped by leading teams.
Each of these can be built and shipped with Runtype — from chat assistants to automation agents.
Top 3 this week:
Shopify
AI merchant assistant that analyzes sales, manages inventory, and answers store questions.
Notion
Autonomous agents that triage tasks, summarize standups, and manage workspace knowledge.
$842K
3 deals
HubSpot
AI data agent that surfaces insights, flags risks, and answers questions across your CRM.
Go to Settings → Security → Reset. I've sent a link to your email.
Intercom
AI chatbot that resolves customer issues instantly using your docs and product data.
Slack
AI recaps of channels and threads so teams stay aligned without reading everything.
Klarna
AI-powered product discovery that compares options and recommends purchases.
Vanta
AI that auto-answers security questionnaires using your compliance knowledge base.
Loom
Automatic AI summaries, chapters, and action items from recorded videos.
Asana
AI teammates that organize workflows, assign tasks, and identify bottlenecks.
Linear
AI that auto-labels, prioritizes, and routes issues based on historical patterns.
The team has been working on
the onboarding experience
for new users across all
Grammarly
Inline AI that rewrites, corrects, and adapts tone while preserving your voice.
AI visual discovery that matches photos to shoppable products.
Instacart
AI search personalized to dietary preferences and shopping habits.
Brands referenced for the kinds of AI experiences you can ship: Shopify (Commerce Copilot); Notion (Knowledge Agents); HubSpot (CRM Intelligence); Intercom (Support Agent); Slack (Thread Summaries); Klarna (Shopping Assistant); Vanta (Security Automation); Loom (Video Summaries); Asana (Task Agents); Linear (Issue Triage); Grammarly (Writing Assistant); Pinterest (Visual Search); Instacart (Smart Search).
Native AI experiences, wherever your users are.
A Surface is one deployed channel — like a chat widget on your site, a Slack bot, or an API endpoint. Every plan includes unlimited Surfaces, each handling unlimited users and conversations. You're billed by usage, not by how many channels you deploy.
Web Chat
Deploy chat widgets, copilots, and conversational AI to any website or app.
Auto-respond to emails, process inbound messages, and draft replies with AI.
API
HTTP endpoints for backend automation, integrations, and data pipelines.
MCP
Extend Cursor, Claude, and ChatGPT with your own tools and data via MCP.
Agent-to-Agent
Let agents delegate tasks to each other. Research, write, review in coordinated handoffs.
Schedule
Cron-based execution for daily reports, data syncs, and periodic monitoring.
Slack
Ship AI bots directly into Slack for support, standups, and team workflows.
Confirmed! Dr. Lee at 3:00 PM today. Reply CHANGE to reschedule.
SMS
Appointment reminders, order updates, and two-way conversations over text.
Webhook
Trigger flows from any external event. Stripe payments, GitHub pushes, form submissions.
Your order shipped today! Tracking: 1Z999...
✓✓ 2:14 PMCustomer support, order tracking, and appointment scheduling over WhatsApp.
Discord
Community FAQ bots, moderation assistants, and server management agents.
3 deploys, 0 incidents. CPU avg 42%.
4:32 PMTelegram
Deploy bots for ops alerts, team commands, and automated reporting in Telegram.
Run it on your own cloud
Run fully managed, or deploy the Runtype runtime to your own infrastructure — AWS, GCP, Vercel, or sovereign clouds. White-labeled, multi-tenant, with SSO and audit trails. Everything you built carries over: agents, flows, tools, and surfaces.
Enterprise
Custom pricing
For companies embedding Runtype into their own product, or running it at scale on their own infrastructure.
For companies embedding Runtype into their own product, or running it at scale on their own infrastructure.
Contact salesFrequently asked questions
What drives the bill
How teams model, measure and reduce the cost of an AI feature once real customers are using it.
How to cut LLM spend without making the product worse
Rank LLM cost optimization by risk to output: measure first, trim context, cache the prefix, cap loops, route by task class, then downgrade the model.
ExplainerHow to price an AI feature when your cost is variable
Seat pricing assumes flat marginal cost. Five ways to price an AI feature when it does not: bundled, seat uplift, credits, usage-based, outcome-based.
GuideHow to stop one customer's usage from destroying your margin
How to limit AI usage per customer: rate limits versus spend limits, per-tenant and per-user budgets, run-level caps, what to do at the cap, and alerting.
GuideHow to work out what one AI conversation actually costs you
How to turn one provider invoice into cost per turn, per conversation, per customer and per resolved task, with the retries, tools and evals included.