Simple pricing.
No per-token markup.
Pay only for the platform. Your provider API costs go directly to OpenAI, Anthropic, or Google — we never touch that bill.
For personal projects and evaluation. No credit card required.
- ✓ 10,000 requests / month
- ✓ 1 workspace
- ✓ 3 API keys
- ✓ AI router + circuit breaker
- ✓ Prompt registry (unlimited prompts)
- ✓ Semver versioning
- ✓ Basic usage dashboard
- ✓ Community support
- ✗ Policy engine
- ✗ Evals
- ✗ GitOps pipeline
- ✗ SSO / SAML
For teams shipping AI features to production.
- ✓ 500,000 requests / month
- ✓ 10 workspaces
- ✓ Unlimited API keys
- ✓ AI router + circuit breaker
- ✓ Prompt registry + GitOps CI/CD
- ✓ Full observability (latency, cost, errors)
- ✓ Policy engine (PII, model, spend, time)
- ✓ Evals (pass/fail, JSON schema, regex, semantic)
- ✓ Budget alerts + quota enforcement
- ✓ Email support
- ✗ SSO / SAML
- ✗ On-premise deployment
For organisations with high volume, compliance requirements, or on-prem needs.
- ✓ Unlimited request volume
- ✓ Unlimited workspaces & API keys
- ✓ SSO / SAML integration
- ✓ Custom RBAC roles
- ✓ Audit log export (S3 / GCS)
- ✓ Custom data retention policy
- ✓ Private cloud or on-premise deploy
- ✓ 99.99% uptime SLA
- ✓ Dedicated Slack support
- ✓ Volume-based pricing
- ✓ Custom contract & invoicing
- ✓ Security review & pen-test reports
Compare all features across plans.
| Feature | Free | Team | Enterprise |
|---|---|---|---|
| Gateway | |||
| Monthly requests | 10,000 | 500,000 | Unlimited |
| Providers (OpenAI, Anthropic, Gemini, Bedrock, Mistral AI, Cohere, Groq, DeepSeek) | ✓ | ✓ | ✓ |
| Circuit breaker & automatic failover | ✓ | ✓ | ✓ |
| Streaming (SSE) | ✓ | ✓ | ✓ |
| Per-key rate limiting | ✓ | ✓ | ✓ (custom) |
| Overage rate (per 1K extra requests) | Blocked | $0.10 | Negotiated |
| Prompt Registry | |||
| Unlimited prompts | ✓ | ✓ | ✓ |
| Semver versioning | ✓ | ✓ | ✓ |
| Per-env deployments (dev/staging/prod) | ✓ | ✓ | ✓ |
| One-step rollback | ✓ | ✓ | ✓ |
| GitOps CI/CD pipeline | — | ✓ | ✓ |
| GitHub webhook integration | — | ✓ | ✓ |
| Evals & Quality | |||
| Eval suites | — | ✓ | ✓ |
| Assertion types (pass/fail, JSON schema, regex, semantic) | — | ✓ | ✓ |
| gate_publish (block deploy on eval failure) | — | ✓ | ✓ |
| Policy & Governance | |||
| Policy engine | — | ✓ | ✓ |
| PII detection & blocking | — | ✓ | ✓ |
| Budget alerts & quota enforcement | — | ✓ | ✓ |
| Security & Access | |||
| Tenant workspaces | 1 | 10 | Unlimited |
| RBAC roles | ✓ | ✓ | ✓ + custom roles |
| SSO / SAML | — | — | ✓ |
| Audit log export | — | — | ✓ |
| Deployment & Support | |||
| Managed cloud | ✓ | ✓ | ✓ |
| On-premise (supported) | — | — | ✓ |
| Uptime SLA | — | 99.9% | 99.99% |
| Support | Community | Email (48h) | Slack (4h P1) |
Common questions about NEXUS plans.
Does NEXUS add a markup on provider API costs?
No. You supply your own provider keys for supported providers such as OpenAI, Anthropic, Google Gemini, AWS Bedrock, Mistral AI, Cohere, Groq, and DeepSeek. Provider costs are billed directly by the provider. The NEXUS subscription covers the platform layer: routing, registry, observability, policy engine, and release management.
What counts as a "request"?
One call to POST /v1/chat/completions counts as one request regardless of token count, model, or whether it uses streaming. A request that fails at the circuit breaker (before reaching the provider) also counts.
What happens when I hit the free tier limit?
Requests above 10,000/month return HTTP 429 (quota exceeded) with a QUOTA_EXCEEDED error code. Your existing deployments and prompt data are unaffected — only new gateway requests are blocked. Upgrade to Team to restore access immediately.
Can I self-host NEXUS for free?
Yes. The core platform is open-source on GitHub. You can run it on your own Postgres + Redis infrastructure without any license fee. Self-hosted deployments don't count toward the managed cloud limits. The Enterprise plan adds a supported on-premise tier with a deployment runbook and SLA-backed support.
Is there an annual discount?
Yes — annual Team plans are billed at $39/month ($468/year vs $588/year monthly). Enterprise annual contracts include additional volume discounts. Contact sales for custom pricing.
Start building today
Free tier. No credit card. Up and running in under 5 minutes.