Deployment patterns
Deployment patterns
Enterprise platform teams evaluate Kamiwaza by deployment pattern, not by vertical. These are the three patterns that cover most gateway configurations.
Use case patterns
Private LLM deployment
Route regulated data to Llama on-prem. Keep sensitive workloads air-gapped from third-party APIs. Enforce PII routing at the gateway level.
Read patternMulti-tenant SaaS
Give each customer their own model policy without spinning up multiple gateways. Per-tenant allowlists, rate limits, and audit buckets.
Read patternFederated inference
Blend private GPU capacity with Bedrock burst. Automatic failover when latency budget is exceeded. Real-time p95 monitoring drives switching.
Read pattern