Simple, transparent pricing
Pay provider prices with zero margin on tokens. Scale up to enterprise when you need governance, compliance, and dedicated support.
Pay-as-you-go
Zero markup on inference. $5 free credit on signup.
- Provider prices, zero margin on tokens
- Smart routing across providers
- Per-user limits and quotas built in
- EU / US data & inference residency
- Auto top-up with configurable thresholds
- OpenAI-compatible API
- Usage analytics & cost tracking
- Community support
A 5% processing fee applies to credit top-ups to cover payment costs. Inference is billed at exact provider prices.
Enterprise
For regulated environments, large volumes, and custom deployments.
- Everything in Pay-as-you-go
- SSO, SAML & SCIM provisioning
- SOC 2, HIPAA, signed DPA & BAA
- Multi-cloud auto-failover
- Dedicated regions & private deployments
- BYOK (bring your own provider keys)
- 99.95% SLA + named CSM
- Custom contracts & annual invoicing
Volume discounts • Annual contracts • Dedicated onboarding
Private Cloud Infrastructure
All inference runs on Amazon Bedrock and Azure OpenAI — enterprise cloud deployments where frontier labs cannot access your data.
Zero Training
Your prompts and completions are never used to train AI models
No Logging
Frontier labs (Anthropic, OpenAI) never see or store your conversations
Enterprise Grade
SOC 2 and HIPAA compliant infrastructure with contractual DPAs
Popular Models
View all models| Model | Provider | Input | Output | Context | EU |
|---|---|---|---|---|---|
anthropic/claude-opus-4.7 | Anthropic | $15.00 | $75.00 | 200K | Yes |
anthropic/claude-sonnet-4.6 | Anthropic | $3.00 | $15.00 | 200K | Yes |
anthropic/claude-haiku-4 | Anthropic | $0.80 | $4.00 | 200K | Yes |
openai/gpt-5.5 | OpenAI | $5.00 | $15.00 | 256K | Yes |
openai/gpt-5.5-mini | OpenAI | $0.60 | $2.40 | 256K | Yes |
openai/o4-mini | OpenAI | $1.10 | $4.40 | 200K | Yes |
google/gemini-2.5-pro | $2.50 | $10.00 | 2M | No | |
google/gemini-2.5-flash | $0.15 | $0.60 | 1M | Yes | |
meta-llama/llama-4-maverick | Meta | $0.20 | $0.60 | 1M | Yes |
meta-llama/llama-4-scout | Meta | $0.17 | $0.85 | 512K | Yes |
mistralai/mistral-large-3 | Mistral | $2.00 | $6.00 | 128K | Yes |
Prices shown per 1M tokens, billed at exact provider rates. EU column indicates availability in the EU region for data residency requirements. All models run on Amazon Bedrock or Azure OpenAI.
Pricing FAQ
- Why is there a 5% fee on top-ups?
- Stripe and card networks charge us processing fees on every deposit. We pass through a flat 5% so you always know what you're paying — no surprises, no minimums. Inference itself is billed at exact provider prices with zero markup.
- Do I need a credit card to try?
- No. Sign up and get $5 in free credits — no card required. The 5% top-up fee only applies when you add more credits later.
- Can I bring my own provider keys?
- BYOK is available on the Enterprise plan. Use your existing Anthropic, OpenAI, or Bedrock contracts and keep Relai's governance, routing, and quota layer on top.
- What happens when I run out of credits?
- Enable auto top-up to keep your apps running. Set a balance threshold and a top-up amount — we'll recharge automatically and email you a receipt.