OpenRouter Pricing in 2026: Plans, Limits and Real Costs
OpenRouter has a $0 Free tier capped at 50 requests per day, a Pay-as-you-go plan with no minimums and a 5.5% platform fee, and an Enterprise tier quoted via sales. Most solo founders should start on Pay-as-you-go and move to Enterprise only when SSO or invoicing becomes required.
Verified 2026-08-21 · official pricing page · OpenRouter in the directory
Quick pricing verdict
- Starting price
- $0 Free tier; Pay-as-you-go requires buying credits (5.5% platform fee on top of token pricing); Enterprise is Contact Sales with no published rate.
- Free plan
- Yes — 25+ free models, 4 free providers, chat and API access, 50 reqs/day and 20 reqs/min rate cap, Community Support.
- Best plan for most people
- Pay-as-you-go — no minimum spend, no lock-in, 500+ models, 80+ providers, email support, high global rate limits, auto-routing and prompt caching.
- Plan to avoid first
- The Free tier — the 50 reqs/day cap runs out quickly for real builds, and failed attempts still count against the daily quota.
- Biggest hidden cost
- The 5.5% platform fee on every Pay-as-you-go credit purchase, applied on top of provider token rates.
- Pricing checked
- 2026-08-21, against the official page
OpenRouter sells access to 500+ models from 80+ providers through one API. Pricing is split across three tiers: Free, Pay-as-you-go, and Enterprise, with token rates set per model and no markup above provider list prices.
The Free and Enterprise plans do not have published flat prices; only the Pay-as-you-go platform fee (5.5%) is stated as a number. Enterprise pricing depends on volume, prepay credits, and annual commits negotiated with sales.
Plans compared
| Plan | Price | Best for | Key limits |
|---|---|---|---|
| Free | $0 | Trying the API on 25+ free models with no payment setup. | 50 reqs/day, 20 reqs/min; 25+ free models and 4 free providers only; Community Support; failed attempts count toward daily quota. |
| Pay-as-you-go | Buy credits (5.5% platform fee) | Solo founders and small teams that want no minimums and access to all 500+ models. | 5.5% platform fee on credit purchases; no minimum spend or lock-in; 1000 request limit on free models with 20 RPM; High global rate limits on paid models. |
| Enterprise | Contact Sales | Teams that need SSO/SAML, contractual SLAs, invoicing, BYOK, regional routing, and dedicated limits. | $25,000 of list price inference/month with no fees, 5% fee after; $200,000 of list price inference/month with no fees, 5% fee after; pricing based on volume commitments and prepayment. |
Source: OpenRouter pricing page, checked 2026-08-21.
What each plan includes
Free
- Includes
- Chat and API access; 25+ free models; 4 free providers.
- Missing
- Activity Logs & Export; auto-routing and preferred vendor selections; Budgets & Spend Controls; Prompt Caching; Management API key; Admin Controls; Data Policy-Based Routing; BYOK; email support; payment options.
- Avoid if
- You are running anything beyond a hobby script — the 50 reqs/day limit and lack of email support will block you.
- Verdict
- Useful only for initial evaluation; failed requests still count toward quota, so reliability experiments are limited.
Pay-as-you-go
- Includes
- 500+ models; 80+ providers; Activity Logs & Export; auto-routing and preferred vendor selections; Budgets & Spend Controls; Prompt Caching; Management API key; credit card, crypto & more payment options; Email Support; High global rate limits.
- Missing
- Admin Controls; Data Policy-Based Routing; Model & provider policies; Managed Policy Enforcement; SSO/SAML; Contractual SLAs; invoicing; BYOK; Shared Slack support channel.
- Avoid if
- You need SSO, formal SLAs, invoicing, BYOK, or regional routing — those are Enterprise-only.
- Verdict
- The right default for most buyers: no minimums, no lock-in, and the 5.5% fee is the only added cost over provider rates.
Enterprise
- Includes
- Everything in Pay-as-you-go plus Admin Controls; Data Policy-Based Routing; Model & provider policies; Managed Policy Enforcement; SSO/SAML; Contractual SLAs; invoicing; BYOK; Optional dedicated limits; Support SLA with Shared Slack Channel.
- Missing
- A public price list — everything is quoted against volume, prepayment credits, and annual commits.
- Avoid if
- You are pre-revenue or under ~$10k/month of inference — the sales process and commitments are not worth it.
- Verdict
- Necessary once SSO, procurement, or compliance land on your roadmap; the fee structure offers a free inference window before the 5% kick-in.
Add-ons, overages, and surprise costs
| Cost item | Trigger | Published amount | Billing behaviour |
|---|---|---|---|
| Platform fee on credits | Every Pay-as-you-go credit purchase | 5.5% | Added on top of whatever you pay the model provider per token. |
| Failed request quota usage on Free | Failed or rate-limited free-tier requests | Counts toward 50 reqs/day | Failed attempts still count toward the daily quota, so outages cost you headroom. |
| Enterprise fee after free window | Enterprise usage beyond the published free tier of inference | 5% fee after $25,000 or $200,000 of list price inference/month (per the two Enterprise fee rows shown) | Applied once monthly list-price inference crosses the threshold; the rate is per the page, not derived. |
| Mid-flight price changes | A model's provider changes its pricing | New provider rate | OpenRouter continues routing to the model and charges the new rate; credits deduct accordingly. |
Source: Platform pricing docs, checked 2026-08-21.
Which plan should you choose?
- Solo founder prototyping an AI feature
- Pay-as-you-go — No minimums, no lock-in, full access to 500+ models and auto-routing, with email support once you need it.
- Hobbyist or student just exploring the API
- Free — $0 entry to 25+ free models via chat and API, but the 50 reqs/day cap and lack of email support will constrain any real workload.
- Startup at ~$5k–$50k/month of inference with security review coming
- Enterprise — Adds SSO/SAML, contractual SLAs, invoicing, BYOK, regional routing, and dedicated limits that a security review will ask for.
- Team that needs procurement-friendly billing today
- Enterprise — Only Enterprise supports invoicing, POs, and SSO; Pay-as-you-go accepts credit/debit, crypto, and bank transfers but no formal invoicing.
Frequently asked questions
- Does OpenRouter mark up provider token prices?
- No. Per the page, OpenRouter does not mark up provider pricing — what is shown in the model catalog is exactly what you pay, plus the 5.5% Pay-as-you-go platform fee.
- Are failed or fallback requests billed?
- No. When routing/fallback is enabled, you are billed only for the successful model run.
- Is there a minimum spend on Pay-as-you-go?
- No. Pay-as-you-go has no minimums and no lock-in — you pay only for what you use.
- What payment methods are accepted?
- Pay-as-you-go accepts credit/debit cards, crypto, and bank transfers. Enterprise supports invoicing and POs; contact Sales for procurement options.
- Are there platform rate limits on paid plans?
- No. There are no platform-level rate limits for Pay-as-you-go or Enterprise users; only Free users are capped (50 reqs/day, 20 reqs/min).
What we read
Every figure on this page was extracted from the vendor's own pricing page at the moment below. The text is reproduced so you can check our reading against what they publish today.
Source captured 2026-08-21 07:12 UTC
Gemini 3.7 Flash is 75% off for a limited time. See all discounted models here (opens in new tab) Dismiss Skip to content Search ⌘ K Models Benchmarks Chat Rankings Apps Enterprise Docs Sign Up Pricing Plans for indie hackers, AI native startups, and enterprises Get Started Talk To Sales Free Pay-as-you-go Enterprise Platform Fees N/A 5.5% Fee discounts available Models Explore all models → 25+ free models 500+ models 500+ models Providers Explore all models → 4 free providers 80+ providers 80+ providers Chat and API Access Try chat now → Activity Logs & Export Auto-routing, preferred vendor selections Learn more → Budgets & Spend Controls Prompt Caching Learn more → Management API key Learn more → Admin Controls Enterprise features → Data Policy-Based Routing Model & provider policies Managed Policy Enforcement SSO/SAML Contractual SLAs Payment options Credit card, crypto & more Invoicing options BYOK Limits Learn more → $25,000 of list price inference / month with no fees, 5% fee after $200,000 of list price inference / month with no fees, 5% fee after Rate limits 50 reqs/day High global limits Optional dedicated limits Token Pricing Free models only No minimum spend. Prices based on models Volume commitments. Prices based on models Support Community Support Email Support Support SLA with Shared Slack Channel Get Started For Free Buy Credits Contact Sales Frequently Asked Questions Billing and Pricing How are tokens billed? Do you mark up provider pricing? How is billing structured for BYOK, Pay‑As‑You‑Go vs Enterprise? Are failed or fallback attempts billed? Do you offer volume discounts or annual plans? Are streaming responses billed differently? What payment methods do you accept? Are taxes (VAT/GST) included in prices? Is there a minimum spend or lock‑in on Pay-as-you-go ? Usage and Rate Limits Do you enforce rate limits? Can I separate environments (dev/staging/production)? Do you enforce platform rate limits? Routing and Latency Can I make sure to send API requests in specific regions? Does routing affect latency? What happens if a model is deprecated or pricing changes? Can I pin specific model versions? Privacy and Security Do you train on customer data? Do you support SSO? Models and Features How do I migrate from OpenAI/Anthropic? Do you support function calling/tools? Reliability and Uptime What happens if a provider is down or a model errors? Where can I check uptime and incidents? Ready To Get Started? Join thousands of developers building with OpenRouter Sign Up For Free Contact Sales © 2026 OpenRouter, Inc Product Chat Rankings Benchmarks Apps Discover Models Collections Providers Pricing Enterprise Labs Company About Blog Careers Hiring Privacy Terms of Service Support Works With OR Data Developer Documentation API Reference Developer Platform Status Connect Discord GitHub LinkedIn X YouTube Cookie Preferences We use essential cookies for site functionality and optional cookies for analytics. Privacy policy Decline Accept STRUCTURED PRICING DATA: {"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How are tokens billed?","acceptedAnswer":{"@type":"Answer","text":"Input and output tokens are billed per model at posted rates."}},{"@type":"Question","name":"Do you mark up provider pricing?","acceptedAnswer":{"@type":"Answer","text":"We do not mark up provider pricing. Pricing shown in the model catalog is what you pay which is exactly what you will see on provider's websites."}},{"@type":"Question","name":"How is billing structured for BYOK, Pay‑As‑You‑Go vs Enterprise?","acceptedAnswer":{"@type":"Answer","text":"Pay-as-you-go: You buy credits and use them as you wish. You can automatically top-up your account or do it manually. You can see the activity in your settings > API Keys Enterprise: Pricing is based on volume, prepayment credits, annual commits, and many other factors."}},{"@type":"Question","name":"Are failed or fallback attempts billed?","acceptedAnswer":{"@t
Pricing change log
What moved since we started tracking this page, newest first. A price that changes silently is the failure mode of every other pricing guide.
- 2026-08-21First published, extracted from https://openrouter.ai/pricing.