Skip to main content

​
Account

Sign up at runinfra.ai/sign-up with Google, GitHub, or email and a password of at least 8 characters. No card required, and new accounts start with $1 in credit for testing the Model APIs. Confirmation and password-reset links work in any browser, and an expired link says so with a resend option.
A workspace groups your API keys, members, and credits, with one balance for Model APIs usage. Every account starts as Personal.
Every account starts as Personal. Only the owner can switch Account type to Business at Settings > Account and enter the company name in Account name. The switch is reversible. Invoices carry the company name, billing address, and tax ID you set at checkout or through Manage billing details at the top of Settings > Billing. A Business account shows the company name, billing address, and tax ID its invoices carry as read-only details on the same Settings > Account page, where Edit in billing opens the billing portal. Audit logs are available on Business accounts and Enterprise plans at Settings > Security. On a Business account, the owner can review every action recorded in the account. The audit log records access, key, member, and billing events from the moment the account becomes a Business account, and nothing before it. Add teammates at Settings > Members.
The owner can enter one email address as the Billing contact at Settings > Account, right after Account name. Purchase receipts, balance and spend alerts, payment issues, plan changes and account holds go here, with the owner copied. The invoice for a purchase follows the payment account, and this setting cannot move it. Clear the address to send those notices to the owner alone, as before. A billing contact gets no account access, sign-in, or permissions. Security and account emails, such as sign-in links and API key expiry warnings, always go to the owner and never to the billing contact.
Yes. RunInfra arranges Model API prices per account and per model. You see your agreed prices while signed in. They apply automatically when you call a covered model with your account’s API key. The response’s usage.cost reflects that price.Partner pricing at Settings > Billing lists each covered model with Your price beside List price. The section is hidden for accounts without arranged prices. GET /v1/models returns your account’s prices when called with its API key.Partner pricing is not self-serve. Contact us.
Open Settings > Members and enter their email. They receive an email invite, valid for 14 days, whether or not the address already has a RunInfra account. They join the inviting account when they select Accept invite in that email and then Join, or when they sign in or sign up with that address. Until then, the owner sees a Pending row and can select Revoke on the same page. There are no per-seat fees.
Yes, self-serve, at Settings > Account. The dialog shows exactly what will be removed, including any remaining credit balance, which is forfeited, and lists any blockers before you confirm. You confirm by typing the confirmation phrase. Password accounts re-enter their password. OAuth-only accounts must have authenticated within the last 10 minutes and can use an email verification code when the session is older. Account deletion waits while an automatic recharge is settling: try again in a few minutes, and contact support if it keeps waiting. Starting deletion stops new automatic recharge attempts while deletion is in progress. An unfinished deletion keeps automatic recharge paused until you finish the deletion or save your auto recharge settings again, which resumes it for every workspace where it is enabled. Deletion is immediate and permanent. Account deletion ends the plan at once, with no refund of the current month.
For a workspace data export, contact support. We can prepare a signed export without exposing API keys.
Two switches at Settings > Notifications.Product notifications (on by default) cover things worth acting on: low or depleted credits, budget alerts, and coding plan limit alerts.Daily billing digest (off by default) summarizes the previous day’s credit purchases and settled usage. Nobody receives it unless they turn it on.Security and account emails, such as sign-in and password links and API key expiry warnings, are always sent to the owner. Preferences are per user and apply immediately.

​
Coding plans

At Settings > Coding plan, where an owner subscribes, changes or cancels it.
At our standard per-token prices, with cached input at the cached price. Plan usage is not a credit deposit and does not roll over. Monthly usage is approximate: months hold different numbers of weekly resets.
Your 5-hour window starts with your first plan request, and the next request after it ends starts a new one. The weekly limit resets on a fixed schedule from activation. The plan pays while both have room.
Credits pay if you chose Keep full speed on credits (with a cap, or no limit), then free Standby serves chat. An owner chooses when subscribing and can change it at Settings > Coding plan. If neither can serve a request, it stops with the reason, reset time and next step.
Free, slower service when we are busy, up to a daily amount. Standby covers chat only. It refills at midnight UTC, may be reduced or paused, and is off while a plan payment is due or processing, or credits are owed.
Chat (chat completions, messages and responses), plus embeddings and rerank whenever those models are available, from every workspace key except keys marked Credits only. Images and audio always stay pay as you go.
Yes, it is shared by the workspace, not per member. Only an owner can buy or change it, or connect a terminal with runinfra login. Members create their own API keys at Settings > API keys and point their agents at them.
Yes. An owner marks its key Credits only at Settings > API keys. It always uses credits and keeps its wallet and key limits.
Requests that fail before returning output never use your plan. Client cancellations count usage, as in pay as you go. Metering changes come with written notice, and promotional rates are labeled temporary.
At the start of each billing period. Your plan renews monthly until canceled. Upgrades apply now with a prorated charge, and counted usage carries over. Downgrades apply at the period end.
After a failed renewal, your plan can keep serving for up to 48 hours, or 7 days if it has had a paid renewal. Select Pay overdue amount at Settings > Coding plan. A disputed payment pauses the plan and ends it at the period end.
Anytime at Settings > Coding plan. Your plan ends at the period end, or at once while a payment dispute is open. Deleting your account ends it at once, with no refund of the current month. Afterwards your keys keep working on credits, which keep their expiry dates. Before your plan ends, Keep plan undoes the cancellation, except after a payment dispute, even one decided in RunInfra’s favor.
A coding plan that never starts serving is refunded in full. A renewal or an upgrade charged after you delete your account is refunded too. See Terms of Service, section 7.
See Coding plan for tiers and the funding API for request receipts.

​
Rate limits

Three limits govern your Model API requests:
  1. Requests per minute, per API key. The default is 10,000 before your first credit purchase and 50,000 after it. Enterprise plans default to 200,000. Provider partners have a provider-grade rate. Each key follows your workspace default. A key that already has a custom limit keeps it, clamped to the workspace maximum.
  2. Tokens per minute, per model. Each model has its own budget for your account. Credit purchases and net coding plan payments contribute to that budget. The credit milestones are below. A model keeps its own default if it is higher than the tier’s minimum.
  3. Concurrent requests, per model, on pay as you go. Before your first purchase, your share is a quarter of the model’s capacity. After your first purchase, it is most of that capacity. These shares apply only while a model is busy. With spare capacity, your account can exceed its share.
The Rate limits section at Settings > API keys shows all three limits and your token tier. The per-model table and the token tiers are under All limits and tiers. Each model’s published concurrency allowance is available on GET /v1/models as max_concurrent_requests_per_api_key.
Add credit at Settings > Billing. Your first credit purchase, from $5, unlocks paid standing with a higher request rate and busy-model concurrency share. Your token budget follows lifetime purchases, not your remaining balance:A top-up raises your token tier as soon as payment settles. Your earned tier never goes down. Promotional credit does not count. Pro and Team coding plans also have a temporary floor while serving covered requests. See Coding plans and Standby.Enterprise is arranged with RunInfra, not unlocked by a purchase. It has no practical token limit, a provider-grade request rate, and the same concurrency share as the top paid tier. Contact us.

​
API keys

Create a workspace API key at Settings > API keys.
Use a workspace API key for hosted models. Select the model with the model field in the request body. See Authentication.
Yes. Open the key’s row menu on the API Keys page and choose Revoke or Rotate. A revoked key is refused on its next request. Requests already in flight can finish.
Yes. On Settings > API Keys, use the copy button on the key’s row. The member who created the key and the workspace owner can copy it; other members and viewers cannot, and every copy is recorded in the audit log. A key created before copying was available was stored only as a one-way digest and cannot be shown again: create a new key, or rotate this one from its row menu, to get one you can copy again. After a rotation the old key keeps working for up to 24 hours while you switch.
Only if you set an expiration when you create the key. You can choose 30, 60, 90 days or 1 year. After a rotation, the previous key keeps working for up to 24 hours, as described in Authentication. An expired key returns 401.
Sign-in today is email and password, or Google or GitHub OAuth. SAML or OIDC single sign-on is not available today. It is an Enterprise conversation. Contact sales.

​
Security

No. Request and response content is never used for training. Prompt text is not stored at all: what we keep per request is usage metadata for billing, and a response body only inside an idempotent replay window you open yourself. The full account, including what exists only while a request is in flight, is in Data retention.
Yes. TLS 1.2 or above everywhere. All traffic to the RunInfra API is over HTTPS.
RunInfra is SOC 2 Type 1 and Type 2 attested. The reports are available to procurement teams on request through the RightNow AI trust center; RunInfra is part of RightNow AI.RunInfra is not offered for protected health information today. HIPAA and a BAA are not available. GDPR commitments, signed data processing terms, region-locked deployments, and VPC or private-link networking are contract conversations rather than self-serve settings. Contact sales.

Pricing and credits

One balance, top-ups, and what happens at zero.

The live model list

Ids, context windows, and discoverable limits.

Troubleshooting

When something is broken.