Comptech AI Gateway
One secure, on-prem gateway in front of every AI model, cloud or self-hosted. Applications use one API; you control which models they reach, what data leaves, what it costs and what is logged.
Overview
Every AI call, through one controlled front door
As teams adopt AI, every application ends up with its own provider keys, its own SDK and no shared view of cost, data or risk. Sensitive data can reach cloud models without anyone noticing.
Comptech AI Gateway sits between your applications and every model. It gives developers one OpenAI-compatible API, and gives security, risk and finance teams central routing, guardrails, budgets and audit logs, all running in your own data centre.
Who it's for
Built for regulated organisations
Banks and financial institutions
Keep customer data on on-prem models, enforce guardrails and give auditors a full log of AI use.
Government agencies
One sovereign entry point for AI across departments, with data-residency rules applied centrally.
Enterprises
Give every team access to approved models with budgets, chargeback and single sign-on.
Platform and AI teams
Offer models as a shared service instead of managing keys and SDKs for every project.
Key benefits
What the gateway gives you
One API for every model
An OpenAI-compatible API in front of cloud providers and on-prem model servers.
Data stays where it should
Data-residency routing sends sensitive requests only to on-prem models.
Guardrails on every call
Prompt-injection blocking, PII masking and content policies before and after the model.
Cost under control
Virtual keys, budgets, rate limits and department chargeback reports.
Reliable by design
Load balancing, fallbacks, retries and caching across models and providers.
Full audit trail
Every request logged with tokens, cost, latency and the policy decisions applied.
Explore
The AI Gateway in detail
Models and API
One OpenAI-compatible API, 100+ models, every endpoint type and model aliases.
Routing and reliability
Load balancing, fallbacks, retries, caching and data-residency routing.
Access and cost
Virtual keys, teams, SSO, budgets, rate limits and chargeback.
Guardrails
Prompt injection, PII masking, moderation and secret detection.
Observability
Request logs, metrics, tracing, alerts and audit logs.
Tools and agents
MCP tool gateway, prompt management and provider pass-through.
Deployment
On-prem, Kubernetes, air-gapped, high availability and integrations.
Architecture
Applications, gateway, models
Applications call the gateway with a virtual key. The gateway checks access and budgets, applies guardrails, routes the request to the right on-prem or cloud model, checks the response, and logs everything.
Applications
- Chat assistants
- Agents and RAG apps
- Business applications
- Developer tools
Comptech AI Gateway
- Access and budgets
- Guardrails in and out
- Routing and fallbacks
- Caching
- Logs and audit
Models
- On-prem models on your GPUs
- NVIDIA AI Enterprise models
- Cloud model providers
Sections
Where to find each capability
Models and API
- What it covers
- Unified OpenAI-compatible API, 100+ models, every endpoint type, model aliases
Routing and reliability
- What it covers
- Load balancing, fallbacks, retries, timeouts, cooldowns, caching, data-residency routing
Access and cost
- What it covers
- Virtual keys, teams, roles, SSO, budgets, rate limits, chargeback, approvals
Guardrails
- What it covers
- Prompt injection, PII masking, moderation, banned keywords, secret detection, policies
Observability
- What it covers
- Request logs, metrics, tracing, alerts and audit logs
Tools and agents
- What it covers
- MCP tool gateway, prompt management, provider pass-through
Deployment
- What it covers
- On-prem, Kubernetes, air-gapped, high availability and Comptech integrations
Included
With every Comptech product
On premises or air-gapped
Runs in your own data centre, so prompts and responses stay under your control.
Single sign-on
SAML and OIDC sign-in, with AD/LDAP groups mapped to teams and roles.
Secrets vault
Provider credentials held centrally, never handed to applications.
DC/DR design
Designed for a primary data centre and a disaster recovery site.
Local support
Comptech engineers in Bangladesh for deployment and support.
Pilot and training
A focused pilot on one use case, plus training for platform and application teams.
Questions
Do applications need to change?
Applications that already use an OpenAI-style API usually only change the base URL and key.
Can we use cloud models at all?
Yes, where your policy allows. Data-residency routing keeps sensitive requests on on-prem models.
Does any data leave our premises?
Only requests your policies send to an approved cloud provider. The gateway, its logs and its keys run in your data centre.
Explore
Other Comptech products
Comptech Container Platform
Enterprise Kubernetes with GitOps, a private registry and DC/DR.
Comptech AI Platform
Private generative AI and MLOps on your own GPUs.
Comptech Data Pipeline
Open lakehouse with SQL, pipelines, dashboards and lineage.
Comptech DocuVault
Document management for banks: capture, document AI, workflows and records.
Comptech AIOps
Monitoring, service desk, AI root cause and DR assurance for bank infrastructure.
Comptech VisionVault
On-prem AI video intelligence, transcription and evidence management.
Comptech SecOps
Security for every workload across your Comptech platforms.
Request a pilot
Tell us which applications and models you want to bring under control first. We'll plan an AI Gateway pilot around them.
Related services