No gateway sprawl
One Apinizer API/AI Gateway runtime — not three (REST + WebSocket + AI) running side-by-side, eating cluster resources.
Unified API & AI Gateway Platform
One platform for every API request — including agents.
Apinizer is the unified API & AI Gateway platform. HTTP, gRPC, SOAP, WebSocket, GraphQL — alongside LLM, MCP, and agent-to-agent traffic — through one runtime on your own server. Policies in code. Audit and encryption built in.
- HTTP · gRPC · SOAP · WS · GraphQL
- Three-tier permissions
- Audit at the persistence layer
Everything is200OK:)
Trusted by 100+ banks, ministries, and defense organizations
Trusted by 100+ organizations across banking, government, and defense
- Aselsan
- Havelsan
- Aktif Bank
- Takasbank
- MKK
- Dünya Katılım
- Emlak Katılım
- Ulaştırma Bakanlığı
- Enerji Bakanlığı
The unified gateway
Legacy SOAP and next-gen AI — through one gateway, governed once.
Banks running WSDL services next to gRPC, GraphQL, and AI traffic don’t need a separate gateway per protocol. The same Worker, the same policy pipeline, the same audit aspect — across every era of API.
One Worker — every protocol
- Legacy
SOAP 1.1 / 1.2
WSDL primed at deploy, WS-Security, WS-Addressing
- Legacy
WSDL imports
Auto-convert to OpenAPI; expose as REST or gRPC
- Modern
HTTP / 1.1
Keep-alive, chunked, content compression
- Modern
HTTP / 2
Multiplexing, server push, binary framing
- Modern
REST
Stateless, HATEOAS, content negotiation, versioning
- Modern
GraphQL
Reverse proxy, query / mutation / subscription
- Modern
gRPC
Protocol Buffers, unary + bidi streams over HTTP/2
- Modern
WebSocket
RFC 6455, binary + text frames, WSS (TLS)
- Next-gen
LLM (OpenAI)
Multi-LLM routing across 17+ providers
- Next-gen
MCP
Model Context Protocol — auto-generated MCP servers
- Next-gen
A2A
Agent-to-agent auth + observability + audit
One audit trail
WSDL calls, gRPC streams, and LLM prompts all land in the same Elasticsearch index, queryable side-by-side.
One permission model
Three-tier (System / Project / Team) governs SOAP-as-REST proxies, AI routes, and the Manager UI uniformly.
Many teams, one platform
Independent teams. Shared infrastructure. No stepping on each other.
Apinizer’s Project model lets unlimited teams run their own work side-by-side under shared infrastructure. Each Project has its own URL prefix, its own member list, and its own scope — proxies, policies, connections, credentials, variables, task flows. Server-side enforcement, not just UI hiding.
- Per-project Gateway Path — every proxy lives under
/<project>/... - Same login, different roles per project — Owner in A, Viewer in B
ProjectAuthAspectrejects cross-project access at the framework layer- Project switcher persists in localStorage — operators stay in context across reloads
Project switcher
⌘Kpayments-coreOwner
/payments-core
- onboarding/onboardingManager
- risk-engine/riskDeveloper
- partner-portal/partnerViewer
- compliance-reports/complianceViewer
Owner
Full control of the project — read, write, deploy, and member management.
Manager
Read, write, and deploy. No member management — that stays with the Owner.
Developer
Read and write the project's proxies, policies, connections — no deploy.
Viewer
Read-only — useful for auditors, partner teams, and read-only integrations.
Many environments, one console
Manage dozens of environments from one screen.
Each Environment is a Kubernetes Namespace — isolated CPU, RAM, and network. The same proxy can run version 1.0 in PROD-EU and version 1.1 in PROD-TR at the same time. No second Manager. No duplicate configuration.
Environments — orders project
6 activedev
DEPLOYED- Type
- TEST
- Protocol
- HTTP+WS
- Version
- 1.4.0-rc2
- Cluster
- Managed
staging
DEPLOYING- Type
- TEST
- Protocol
- HTTPS
- Version
- 1.3.4
- Cluster
- Managed
prod-eu
DEPLOYED- Type
- PRODUCTION
- Protocol
- HTTPS
- Version
- 1.3.2
- Cluster
- Remote
prod-tr
DEPLOYED- Type
- PRODUCTION
- Protocol
- HTTPS
- Version
- 1.3.2
- Cluster
- Remote
prod-air
IDLE- Type
- PRODUCTION
- Protocol
- HTTPS
- Version
- 1.3.0
- Cluster
- Remote
perf-lab
DEPLOYED- Type
- TEST
- Protocol
- GRPC
- Version
- 1.4.0-rc2
- Cluster
- Managed
1:1 K8s namespace mapping
Each environment isolates CPU, RAM, network.
TEST or PRODUCTION
Two types drive licensing and resource allocation.
HTTP, HTTPS, gRPC, WS
Four protocol modes per environment.
Per-environment versions
Same proxy, different version, side-by-side.
Variables scope by env
${var} resolves to a different value in PROD vs TEST.
Connections scope by env
Database / Kafka / SMTP credentials per environment.
Hybrid by design
Managed or remote.
One console, both at once.
Some environments live in the cluster Apinizer manages for you. Others run on your own Kubernetes — region-local, sovereign, or air-gapped. Same Manager, same audit trail, same identity, same permissions. The only difference operators see is a small Managed or Remote badge.
Managed
Apinizer-managed cluster
The Apinizer API/AI Gateway runs in the cluster Apinizer operates next to the Manager. Fast iteration, predictable scale, hands-off operations. Ideal for TEST and STAGING.
- Apinizer scales, monitors, and patches the runtime
- Same Manager UI, same APIops surface
- Hot deploy without dropped sockets
Remote
Customer-owned cluster
The Apinizer API/AI Gateway runs on your Kubernetes — EKS, AKS, GKE, OpenShift, Tanzu, RKE2, or on-prem. Sovereign, region-local, air-gapped — the data plane stays where regulation requires it.
- Connects via
apiServerUrl+ service account token - Credentials stored encrypted (
@SecretData) - Air-gapped variant for clusters without external network
Real hybrid topology
TEST managed for fast iteration · STAGING remote on AWS EKS in eu-west-1 · PROD remote on on-prem OpenShift, air-gapped · DR remote on Azure AKS in another region. One Manager. One audit trail. One permission model.
Now with AI Gateway
Govern every LLM, MCP, and agent-to-agent request.
The same gateway that runs your REST APIs now governs every AI request your applications make. One audit trail. One permission model. One place to set the rules.
LLM Governance
Token tracking, cost ceilings, leak prevention. Per user, per model, per window.
Multi-LLM Routing
17+ providers behind one OpenAI-compatible facade. Switch models without code changes.
MCP + A2A
Auto-generated MCP servers, agent discovery, agent-to-agent auth and audit.
Semantic Cache + Prompt Guards
Drop token spend on repeat prompts. Block injection patterns. PII sanitization.
17+ LLM providers — one OpenAI-compatible facade
- OpenAI
- Anthropic
- Bedrock
- Azure OpenAI
- Gemini
- Vertex AI
- Cohere
- Mistral
- Hugging Face
- Llama
- + 7 more
The Apinizer suite
Ten ways to ship safer APIs and AI traffic.
API Gateway and AI Gateway are the data plane. Identity, Analytics, Portal, Designer, Creator, Integrator, Cache, and Monitoring round out the platform — all configured from the API Manager.
CORE PRODUCT
Worker — data plane
API Gateway
Apinizer API/AI Gateway — multi-protocol data plane for HTTP, gRPC, WebSocket, SOAP, GraphQL, and AI traffic on one Java 25 runtime. 50+ policies for auth, traffic, transformation, and observability — out of the box.
NEW
AI Gateway — agentic plane
AI Gateway
Govern every LLM, MCP, and agent request on the Apinizer AI Gateway. Token budgets, cost attribution, response caching, prompt firewalls, and multi-LLM routing across 17+ providers — on the same runtime as your REST APIs.
Identity Manager
Identity Manager
OIDC, OAuth 2.0, JWT, JOSE — plus mTLS, SAML, LDAP / AD, and database-backed federation. One identity surface for every Apinizer API and AI request, with three-tier permissions and Repository-layer audit.
API Portal
API Portal
Open the front door for developers, partners, and AI agents. Stand up as many portals as you need — Public, Internal, Partner, Agents — each branded as your own. No own database, fully customizable, MCP-ready, live in regulated environments.
Analytics Engine
Analytics Engine
Watch every API call. Ship the log everywhere. Answer in seconds. Async traffic capture, fan-out to nine destinations in parallel, dashboards and tracing operators actually use, and anomaly detection with EMA + Bollinger Band — backed by Elasticsearch.
API Designer
API Designer
Design the API. Ship the proxy. One click apart. OpenAPI 3, Swagger 2, and WSDL editing — with a one-click path from spec to a live, governed API on the same gateway, and instant documentation in every format your team needs.
DIFFERENTIATOR
API Creator — three ways to skip the microservice
API Creator
Apinizer API Creator — three ways to ship an endpoint without a microservice. DB to API turns SQL into REST. Mock API gives frontends real-shaped responses with auth. Script to API runs JavaScript or Groovy as an endpoint.
API Integrator · Task Flow Manager
API Integrator
Connect anything to anything. Without writing the integration. Eighteen task types on one drag-and-drop canvas, chained with output keys and JSON Path, fired by cron, HTTP, or another flow — and governed by the same audit and identity stack as the gateway.
Cache
Cache
Apinizer Cache is a two-tier cache — a local in-pod tier inside every gateway, a Hazelcast cluster on Kubernetes for the distributed tier. Twelve first-party cache scopes, atomic invalidation on redeploy, throttle and quota on the same cluster.
Monitoring
Monitoring
Apinizer Monitoring — uptime probes from your regions, anomaly detection with EMA and Bollinger bands, ten platform alert triggers ready on day one, and severity-aware action chains to Slack, Microsoft Teams, ServiceNow, email, webhook, and more.
Deploy anywhere
Kubernetes, OpenShift, Tanzu — or plain VMs.
One platform, four runtimes. Apinizer ships the same artefacts to every target — no fork, no separate edition. Whether your standard is enterprise Kubernetes, Red Hat OpenShift, VMware Tanzu, or traditional virtual machines, the gateway, manager, and APIops surface stay identical.
Supported
Kubernetes
Vanilla, EKS, AKS, GKE, RKE2.
Helm charts · operator-managed · hot deploy without dropped sockets.
Supported
OpenShift
Red Hat enterprise clusters.
SCC-aware manifests · OpenShift routes · air-gapped registries.
Supported
Tanzu
VMware-native Kubernetes.
TKGm / TKGs supported · vSphere CSI · TMC observability hooks.
Supported
Plain VMs
RHEL, CentOS, Ubuntu, Windows Server.
systemd unit · zero container runtime needed · same artefacts as K8s.
Mix and match — one platform
A single Apinizer environment can blend runtimes. Some workers on Kubernetes, others on legacy virtual machines, a third group on OpenShift in another data centre — all under the same Manager, the same identity, the same audit trail. Move at your own pace; the platform doesn't force a rebuild.
- Same artefact layer — no per-platform fork
- One Manager federates K8s + VM workers
- Promote VM-bound services to containers gradually
- Air-gapped + connected workers under the same control plane
On-prem · Sovereign cloud · Public cloud · Air-gapped — same binary
Developer experience
Operate APIs the way your team operates code.
Policies as classes, scripts you can drop in today. Apinizer's surface area is small enough to read end-to-end and explicit enough to review.
orders-api.yaml
# Deploy a proxy from CI — references are by name, not internal ID$ apinizer apply -f orders-api.yaml> proxy: orders-api environment: prod> status: deployed revision: 420+
Organizations on Apinizer
Banks, ministries, defense — across regulated industries.
0
Native protocols
HTTP / HTTP/2 · REST · SOAP · gRPC · WebSocket · GraphQL.
0+
Codified business invariants
Domain rules enforced by the platform — extend with your own custom plugins.
0
Latest release
Four-month cadence — only the latest major receives patches.
Industries
Designed where APIs are the regulated surface.
Banking
Audit, encrypted secrets, and three-tier access for the APIs that touch payments and identity.
Government
Bilingual TR + EN, runs entirely on the customer's Kubernetes — no calls home.
DESIGNED FOR
Telecom
Low-latency API tier in front of OSS / BSS, with hot deploy and async logging.
Energy
API tier in front of SCADA-adjacent systems, EMS, and partner billing — with anomaly detection and certificate watchdogs.
DESIGNED FOR
Insurance
Quote, bind, and claim APIs that need versioning and observability without a rewrite.
Defense
Air-gapped or sovereign clusters with strict change control and zero-trust defaults — no calls home, rootless containers, and an audit trail you can present to oversight.
Built on the standards your team already runs
No vendor lock-in. No surprise dependencies.
Apinizer is a JVM-native platform that fits where Java, Spring, and Kubernetes already live.
Runtime
- Java 25
- Spring Boot 3.5
- Undertow
- MongoDB
- Elasticsearch
- Hazelcast
Frontend & DX
- Angular 19
- PrimeNG 19
- Tailwind CSS
- Groovy
- Quartz
- OpenAPI
Standards
- OAuth 2.0 / OIDC
- WS-Security
- GraphQL
- gRPC
- WSDL
- K8s / OpenShift
Ready when you are
Govern every API — including AI — on your cluster.
A 30-minute walkthrough of API Gateway, AI Gateway, Manager, Portal, and the rest of the suite — on a Kubernetes of your choice.