Free
Explore and prototype
Try it on your own site.
- Your own AI key, zero markup · fair use
- Workspace Architect + test suites
- 1 bot · 3 prompts · 2 guardrails
- 1 knowledge source + 1 connector
- 7-day trace history
Pricing · your key · zero markup
Plans from $0. Your bot runs on your own AI key, so you pay your AI provider directly (typically a few dollars per 1,000 replies on a low-cost model) and Rylvo never marks it up. Start with 14 days of Pro and $2 of AI on us. No card.
No commitment · cancel anytime
Choose your workspace
Every plan uses your own LLM key with zero Rylvo markup — and every new workspace starts with 14 days of Pro and $2 of AI usage included, so you can build before adding one. Upgrade for platform capacity and advanced capabilities.
Explore and prototype
Try it on your own site.
One bot for your own business
Early price. Guaranteed for your first 12 months. $19/mo for new customers from 31 Jan 2027, or after the first 100 customers.
Starter takes no discount or referral codes: the early price is the whole offer.
One bot that answers, captures leads and sends them where they need to go.
Launch a solo product
For builders wiring bots into their own product.
Run the full platform
+ $40/mo per teammate
Everything Rylvo does, for your team or your first clients.
Scale with your runbook
Annual agreements shaped around your volume, security and support
For companies running bots with their own team, on their own AI contracts.
Seats
Pro includes your seat. Every teammate after you is a $40/month seat ($32/month on annual billing) — added or removed whenever you like, prorated the moment you do it, with no plan change and no sales call. Up to 100 seats; past that, talk to us about Enterprise.
What a workspace costs
Monthly billing. Annual billing drops Pro to $80/month and each seat to $32/month.
How billing actually works
Bot chat and AI features like Agent Evolution, Prompt Optimization and Red Team run on your own LLM key. Rylvo adds zero markup, never charges for AI, and has no AI credits. Knowledge-base indexing is included up to your plan's allowance, and the only wallet pays for storage overage and optional add-ons. You pay your provider directly for tokens, exactly as you would anyway.
Meter 1 · Runtime
Every bot chat call uses your own API key — OpenAI, Anthropic, Google, OpenRouter, DeepSeek, xAI, or any compatible provider. Rylvo routes the call server-side and logs a full trace for observability, but adds zero markup on token costs. You pay your LLM provider directly.
0%
markup on AI
no AI credits
Also BYOK · Intelligence
The AI that builds, evolves, and optimizes your bots — Agent Evolution, Self-Improving Prompts, Edge Case Engine, Red Team, Broadcasts — runs on your own key too. Free, with zero markup.
How AI features work
Edge Cases, Agent Evolution, Prompt Optimization, Red Team and Broadcasts all run on your own LLM key — free, with zero markup. No AI credits, no markup. Knowledge-base indexing is included up to your plan's allowance, and the only wallet pays for storage overage and optional add-ons.
Typical token cost per action on your own provider — what you'd pay your LLM provider directly, with zero Rylvo markup. You never pay Rylvo for AI.
Compare plans
Every feature and limit, laid out once. No asterisks, no fine print.
| Capability | Free | Starter | Lite | ProBest for teams | Enterprise |
|---|---|---|---|---|---|
| Build & run | |||||
| Bot chat (BYOK) Bot chat runs on your own LLM key. Rylvo adds no markup and charges nothing per message. Fair use applies; see Bot replies. | Your key, zero markup · fair use | Your key, zero markup · fair use | Your key, zero markup · fair use | Your key, zero markup · fair use | Your key, zero markup · fair use |
| Bots | 1 | 1 | 3 | 25 | Unlimited |
| Prompts | 3 | 10 | 50 | Unlimited | Unlimited |
| Guardrails Automated safety rules enforced before a response is sent. | 2 | 10 | 25 | Unlimited | Unlimited |
| Image & voice intake Customers can send photos, screenshots, documents and voice notes on any channel — the bot reads and hears them. Runs on your own LLM key, so there is no markup. | Included | Included | Included | Included | Included |
| Images read / month Photos, screenshots and document pages processed per billing period. | 100 | 500 | 2K | 20K | Custom |
| Voice minutes / month Audio transcribed per billing period. Rounded up per clip. | 30 | 120 | 600 | 6K | Custom |
| Knowledge sources Docs, files, and URLs your bots retrieve answers from. | 1 | 5 | 15 | 50 | Unlimited |
| Connectors (MCP / webhook) External tools and APIs your bots can call. | 1 | 3 | 15 | 50 | Unlimited |
| Flows (guided & strict) Deterministic step-by-step conversation paths the engine enforces — per-step goals, required data, allowed tools. | 1 | 3 | 10 | Unlimited | Unlimited |
| Beam members People who can sign in to your Members-only Beams, counted across the whole workspace. Invites and access codes stop working once the limit is reached. | 5 | 50 | 500 | 5,000 | Unlimited |
| Test cases per bot | 5 | 10 | 50 | Unlimited | Unlimited |
| Custom test scoring Set your own pass bar, weigh what matters, add score dimensions of your own, and see which tests would flip before you save. | Not included | Not included | Not included | Included | Included |
| Test checks Check a reply for a pattern, JSON of a given shape or a number, and check what the bot did: the tools it called, whether it used the knowledge base, how fast it replied and what it cost. Each check fails the test on its own — no AI judge can overrule it. | Not included | Not included | Not included | Included | Included |
| Judge calibration Your reviewers set their own verdict on judged test results; see how often the AI judge agrees with them (with Cohen's kappa and a confusion matrix), where it is too generous or too strict, and whether a scoring change makes it agree more. | Not included | Not included | Not included | Not included | Included |
| Release gate Hold a release to Production until its tests clear the bar you set — a minimum pass rate, no failing critical test, a named test set, how often the judge agrees with your reviewers — and ask the same gate from your own CI pipeline. | Not included | Not included | Not included | Not included | Included |
| Rules & scoring as code Every save of your detection rules, test scoring and release gate is kept as a version on every plan. Put an earlier one back in a click — as a new version, checked like any save — and export or import them as JSON files. | Not included | Not included | Not included | Not included | Included |
| Fleet policies Lock a detection rule, the test pass bars or a score dimension so every bot follows the workspace's setting. A bot's own change to it is kept but not applied, and a lock already saved keeps holding if you change plans. | Not included | Not included | Not included | Not included | Included |
| AI features (free on your own key) | |||||
| AI features run on your LLM key Edge Cases, Agent Evolution, Prompt Optimization, Red Team and Broadcasts run on your own LLM key — free, zero markup. No AI credits, no markup. Knowledge-base indexing is included up to your plan's allowance. The only wallet pays for storage overage and optional add-ons. | BYOK | BYOK | BYOK | BYOK | BYOK |
| Agent Evolution Reviewer agents analyze your runs and auto-suggest guardrails. | Not included | Not included | Not included | Included | Included |
| Skills Instructions a bot follows only when a customer's message calls for them — with a library of ready-made ones, versions and automatic rollback. | Not included | Not included | Not included | Included | Included |
| Self-Improving Prompts Suggests prompt changes for you to review (OPRO, APE, DSPy and PromptBreeder strategies). | Not included | Not included | Not included | Unlimited | Unlimited |
| Edge Case Engine Finds failure patterns in production and auto-generates guardrails. | Not included | Not included | Not included | Included | Included |
| AI review on automatic scans Every automatic scan also reads recent conversations against your AI review criteria, on your own key, under a cost cap you set. | Not included | Not included | Not included | Not included | Included |
| Detection rules of your own Rules you write over what a conversation records — file a case when the bot promises a refund, names a competitor or takes too long. Tried on recent conversations before you save. | Not included | Not included | Not included | 10 | 100 |
| Red Team testing Adversarial inputs that probe your bot for weaknesses. | Not included | Not included | Not included | Unlimited | Unlimited |
| Your data (BYODB) | |||||
| Storage targets Push every conversation to your own database or webhook. | — | — | 1 | 5 | Unlimited |
| Included events / month Conversation storage events included free. Overage: $0.50 per 1K events. | — | — | 250K | 1M | Custom |
| Observability | |||||
| Trace history How long traces are kept. Extend with retention add-ons. | 7 days | 14 days | 14 days | 90 days | Custom |
| Datasets Curated collections of traces for analysis and review. | 1 | 1 | 10 | 50 | Unlimited |
| Preserved traces Distinct traces protected from auto-deletion across all datasets. | 50 | 200 | 5,000 | 25,000 | Unlimited |
| Trace Replay Step through any past conversation, turn by turn. | Not included | Not included | Included | Included | Included |
| Advanced analytics Rooms of charts on every reply — volume, quality, speed and cost — plus alerts when a number crosses a line. | Not included | Not included | Included | Included | Included |
| Mission Control Real-time oversight of live conversations with intervention. | Not included | Not included | Not included | Included | Included |
| Automation & reach | |||||
| Scheduled tasks Cron-based scans, reviews, and report generation. | 0 | 0 | 10 | 25 | Unlimited |
| Notification channels Slack, email, webhook, Discord, Teams. | 0 | 2 | 10 | 25 | Unlimited |
| Broadcasts Send messages to your audience across WhatsApp, Telegram, Email, SMS, and more. | Not included | Not included | Not included | Included | Included |
| Team & API | |||||
| Team members Free, Starter and Lite are single-seat plans. Pro starts at your seat and adds teammates for $40/seat/month ($32 on annual) — prorated instantly, no plan change. | 1 | 1 | 1 | 1 + $40/seat | 25 seats, more by contract |
| Custom roles & permissions | Not included | Not included | Not included | Included | Included |
| Agency & client workspaces Pro/Enterprise on both sides. The agency sends a role-scoped request; the client owner approves it. Client ownership, data, settings, and billing remain isolated. | Not included | Not included | Not included | Included | Included |
| API keys | 0 | 0 | 10 | 25 | Unlimited |
| Bot replies Replies your bots send, on your own AI key with zero Rylvo markup. Fair use: there is no per-reply charge. Pro's guide figure is about 100,000 replies a month; Enterprise volume is set in your contract. | Fair use | Fair use | Fair use | Fair use · guide ≈ 100K/mo | Set in your contract |
| Security & support | |||||
| SSO (SAML) & SCIM | Not included | Not included | Not included | Not included | Included |
| Custom DPA & security review | Not included | Not included | Not included | Not included | Included |
| Support Enterprise: named contact, next business day (not an SLA). Other paid plans: email, no response target. Free: docs. | Docs | Named contact | |||
| Start for Free | Get Starter for $10 | Start Lite | Start Pro | Talk to Sales | |
Retention add-ons
Keep traces, Mission Control audits, audience data, and synced conversations longer than your plan’s included window — per domain, or all three at once with the discounted Retention Pack. On Starter and Lite they add up to 30 days; longer history is on Pro.
Retention Pack Save 25%
Extend traces, Mission Control audits, audience data, and conversation-storage records together. It is simpler and costs less than combining the three individual add-ons.
Choose a window
Added to the retention already included in your plan.
Extend by
+30 days
$8/ month
$11 separately · save $3
Extend by
+90 days
$19/ month
$26 separately · save $7
Extend by
+180 days
$33/ month
$44 separately · save $11
Extend by
+365 days
$59/ month
$79 separately · save $20
Available on Lite and above. Stack with your plan’s included retention. A pack replaces the individual add-ons (activating one credits back any overlap).
Enterprise
Annual agreements shaped around your volume, security and support: negotiated limits, a custom DPA, a security review with your team, a named contact with a next-business-day target (not an SLA), and SSO / SAML single sign-on against your own identity provider.
Frequently asked
If something’s missing, our team will answer it in writing — just ask.
Free is $0 for one assistant. Starter is $10/month at the early price (then $19 for new customers) for one assistant with no Rylvo branding. Lite is $40/month for 3 assistants. Pro is $100/month for 25 assistants, with agency and client workspaces, and each teammate after you is $40/seat/month. On monthly billing your first month costs less for new customers, once per customer — Lite $25, Pro $80 — and annual billing is about 20% off ($32/month for Lite, $80/month for Pro). Enterprise is custom: an annual agreement shaped around your volume, security and support, so talk to sales for a quote. Every new workspace starts with 14 days of Pro and $2 of AI usage, no card needed; after that, AI usage runs on your own provider key at cost, with zero markup.
Nothing extra. Edge Case Engine, Agent Evolution, Self-Improving Prompts, Red Team, and Broadcasts all run on your own LLM key. No AI credits, no markup. Knowledge-base indexing is included up to your plan's allowance. The only wallet pays for storage overage and optional add-ons. You pay your LLM provider directly for the tokens, exactly like bot chat.
Not to start. Every new workspace gets 14 days of Pro with $2 of AI usage included (once you verify your email), and reading your website or documents into a knowledge base is included on every plan — about 2,500 pages a month on Free and 65,000 on Pro. To keep your assistants answering after the trial, add your own provider key (OpenAI, Anthropic, Google, OpenRouter, and more) in Settings → LLM Providers: you pay your provider directly, with zero markup, and the same key powers every AI feature.
Yes — every new workspace starts on Pro for 14 days, automatically, with no card. It includes $2 of AI usage, so your assistant can answer before you add a key. When the trial ends you move to Free unless you upgrade, and the trial's AI usage ends with it, so add your own provider key to keep your assistant replying. Free includes one assistant: if you have more, you get 7 days to choose which one to keep (or to upgrade), then the extras are paused, never deleted, and when you upgrade they switch back on, up to your new plan's assistant limit. Demo assistants you haven't made your own are removed when the trial ends, and Free keeps each person's data for a shorter time than paid plans.
Never. Your bot's replies run on your own provider key (during your trial, on the AI usage we include). Rylvo routes them server-side and logs a full trace for observability, but adds zero markup. Plans differentiate on features and limits, not AI spend.
Free, Starter and Lite have one person by design: a second person is the moment to move to Pro. Pro is $100/month and includes your own seat. Every teammate is a $40/seat/month add-on ($32/seat/month on annual billing) — added or removed anytime, prorated instantly, with no plan change. Two people cost $140/month, five cost $260/month, and Pro supports up to 100 total seats. Enterprise includes 25 seats, with more agreed by contract.
Pro and Enterprise. Both the agency workspace and the client workspace must be on Pro or Enterprise before managed access can be activated. Lower plans can still view and decline incoming requests, and every plan can end a relationship or revoke access. Agency delegates are the agency's own people: they use the agency's seats, never a seat in the client workspace.
Lite's and Pro's first month is discounted while we're in our launch period, for new customers: Lite is $25 (then $40/mo) and Pro is $80 (then $100/mo). It's applied automatically at checkout on monthly billing — no code needed. The first-month price is for new customers, once per customer: if you have already had a paid Rylvo subscription, on this workspace or another, you pay the regular price from the first month. Seat add-ons are charged at their normal rate. From the second month you pay the regular price.
New customers pay $10/month, guaranteed for their first 12 months. Starter is sold monthly only, with no annual option. On 31 Jan 2027, or after the first 100 customers, Starter becomes $19/month for new customers. If your price ever changes, we tell you at least 30 days ahead. One Starter per business.
Yes. Upgrades take effect immediately and are prorated; downgrades apply at the next billing cycle. Bots and client data are preserved. If moving below Pro, you must first release your purchased seats and deliberately end active or pending agency/client partnerships, so neither a teammate nor delegated access is removed as a surprise. A forced administrative or billing downgrade ends those relationships and revokes only the delegated memberships and API keys.
No AI credits and no markup. An optional prepaid wallet pays only for extra bring-your-own-database storage ($0.50 per 1,000 events above your plan, on Lite and Pro) and other optional add-ons. Knowledge-base indexing is included up to your plan's allowance.
Cards and the other methods Dodo Payments supports. Dodo is our merchant of record, prices are in USD and India GST is added at checkout. Enterprise is billed annually; ask us about invoice terms.
Yes. Email contact@rylvo.com within 14 days of your first payment on a paid plan (monthly or annual) and we refund it in full through Dodo Payments, once per customer, unless the account was suspended for violating the Terms. Renewals, seats, other add-ons and wallet top-ups bought later are non-refundable, except where the law requires or under Dodo Payments' buyer terms. If a subscription payment is refunded in full, the subscription ends and the workspace moves to the Free plan; a partial refund does not end it.
Start with zero risk
Spin up a bot with your own LLM key, trace every call, and upgrade only when you need more workspace capacity and operating features.
Cancel anytime