# Vikat

- [OSS (103 pages)](https://docs.vikat.ai/_llms/oss.md): Documentation for OSS.

## Enterprise

### Getting Started

- [Vikat Enterprise Gateway](https://docs.vikat.ai/enterprise/overview.md): Production-grade AI gateway for organizations running mission-critical AI workloads. Built on top of open-source Vikat with high-availability clustering, fine-grained governance, audit-grade compliance, and managed deployment options.

### Moving from OSS

- [Overview](https://docs.vikat.ai/enterprise/moving-from-oss/overview.md): What carries over from open-source Vikat, what changes, and how to migrate your SQLite config store to PostgreSQL before upgrading.
- [Sizing & redundancy](https://docs.vikat.ai/enterprise/moving-from-oss/sizing.md): Hardware sizing for Vikat Enterprise gateway pods and PostgreSQL, with and without object storage for large logs.
- [Cross-region deployment](https://docs.vikat.ai/enterprise/moving-from-oss/cross-region.md): Why Vikat tolerates geographically distributed pods without putting database latency on the request path.
- [Security hardening](https://docs.vikat.ai/enterprise/moving-from-oss/security-hardening.md): Mandatory controls before exposing Vikat Enterprise to production traffic: IdP-enforced identity, scoped virtual keys, locked-down CORS, and a minimal header allowlist.
- [Versioning](https://docs.vikat.ai/enterprise/moving-from-oss/versioning.md): Why Enterprise and OSS use independent version numbers, and how to find the OSS base each Enterprise release is built on.

### Release Cadence

- [Release Cadence](https://docs.vikat.ai/enterprise/release-cadence.md): How Vikat Enterprise releases are versioned and shipped

### Migration Guides

- [Migrating to Enterprise v1.4.0](https://docs.vikat.ai/enterprise/migration-guides/v1.4.0.md): Breaking changes and migration instructions for the Enterprise v1.4.0 release

### Features

- [User Provisioning (OIDC + SCIM)](https://docs.vikat.ai/enterprise/user-provisioning.md): Authenticate users, sync teams, and provision roles and business units from your identity provider using OAuth 2.0 / OIDC, background directory sync, and inbound SCIM 2.0.
- [Access Profiles](https://docs.vikat.ai/enterprise/access-profiles.md): Define reusable provider, model, budget, rate-limit, and MCP policies that auto-allocate virtual keys to users at scale.
- [Role-Based Access Control](https://docs.vikat.ai/enterprise/rbac.md): Manage user access with fine-grained permissions across Vikat resources using roles and permissions.
- [Sign-in security](https://docs.vikat.ai/enterprise/sign-in-security.md): Second factor, session timeouts, login lockout and the admin IP allowlist.
- [Data Access Control (DAC)](https://docs.vikat.ai/enterprise/data-access-control.md): Restrict row-level visibility of configuration and operational data based on the authenticated caller's role, team, and identity.
- [MCP Tool Groups](https://docs.vikat.ai/enterprise/mcp-tool-groups.md): Curated collections of MCP tools attachable to virtual keys, teams, customers, users, providers, or API keys, and enforced at request time.
- [Audit Logs](https://docs.vikat.ai/enterprise/audit-logs.md): Track administrative activity in Vikat Enterprise with signed audit events, filtering, and export support.
- [Secret Management](https://docs.vikat.ai/enterprise/secret-management.md): Connect AWS Secrets Manager, GCP Secret Manager, or HashiCorp Vault so Vikat never stores plaintext API keys in its database.
- [Log Exports](https://docs.vikat.ai/enterprise/log-exports.md): Offload Vikat request and response payloads to S3 or GCS object storage while keeping searchable metadata in the logs database.
- [Datadog](https://docs.vikat.ai/enterprise/datadog-connector.md): Native Datadog integration for APM traces, LLM Observability, and metrics
- [Running multiple replicas](https://docs.vikat.ai/enterprise/clustering.md): What is shared between gateway nodes, what is not, and what that means for limits.
- [Adaptive Load Balancing](https://docs.vikat.ai/enterprise/adaptive-load-balancing.md): Advanced load balancing algorithms with predictive scaling, health monitoring, and performance optimization for enterprise-grade traffic distribution.
- [Circuit breaker](https://docs.vikat.ai/enterprise/circuit-breaker.md): Stop hammering a provider that is failing, and fall back while it recovers.
- [In-VPC Deployments](https://docs.vikat.ai/enterprise/invpc-deployments.md): Deploy Vikat within your private cloud infrastructure with VPC isolation, custom networking, and enhanced security controls for enterprise environments.

#### Guardrails

- [Guardrails](https://docs.vikat.ai/enterprise/guardrails.md): Content policy for LLM traffic: keyword, regular-expression, PII and model-based rules applied to prompts and to responses.
- [Secrets Detection](https://docs.vikat.ai/enterprise/guardrails/secrets-detection.md): Catch leaked API keys, tokens and private keys in prompts and responses using regex guardrails.
- [Custom Regex](https://docs.vikat.ai/enterprise/guardrails/custom-regex.md): Write organisation-specific content rules with RE2 regular expressions, and use the built-in PII patterns.

### Security

- [Security at Vikat](https://docs.vikat.ai/security.md): Security practices for Vikat: container hardening, dependency scanning, and reproducible builds.

## Integrations

### CLI Agents & Editors

- [Overview](https://docs.vikat.ai/cli-agents/overview.md): Use Vikat with LibreChat, Claude Code, Codex CLI, Gemini CLI, Qwen Code, and more by pointing each tool at the correct Vikat endpoint.
- [Cursor](https://docs.vikat.ai/cli-agents/cursor.md): Add Vikat as a custom model in Cursor, configure MCP tools, and use virtual keys for team access control.
- [LibreChat](https://docs.vikat.ai/cli-agents/librechat.md): Integrate LibreChat with Vikat to access any AI provider through a modern open-source chat interface with virtual keys and observability.
- [Claude Code](https://docs.vikat.ai/cli-agents/claude-code.md): Use Claude Code with Vikat to route through any provider and unlock advanced features like MCP tools and observability.
- [Claude Desktop](https://docs.vikat.ai/cli-agents/claude-desktop.md): Route Claude Desktop App traffic through Vikat for multi-provider routing, virtual keys, and observability.
- [Claude for Office](https://docs.vikat.ai/cli-agents/claude-for-office.md): Use Claude for Office (Microsoft 365 add-in) with Vikat to route requests through any provider with virtual keys, budget controls, and observability.
- [Codex CLI](https://docs.vikat.ai/cli-agents/codex-cli.md): Use OpenAI's Codex CLI with Vikat for powerful code generation with any provider.
- [Gemini CLI](https://docs.vikat.ai/cli-agents/gemini-cli.md): Use Google's Gemini CLI with Vikat for advanced reasoning capabilities with any provider.
- [Qwen Code](https://docs.vikat.ai/cli-agents/qwen-code.md): Use Alibaba's Qwen Code with Vikat for AI-powered coding with any provider, virtual keys, and observability.
- [Opencode](https://docs.vikat.ai/cli-agents/opencode.md): Use Opencode with Vikat to access any AI provider through a terminal-based coding assistant with virtual keys and observability.
- [Zed Editor](https://docs.vikat.ai/cli-agents/zed-editor.md): Integrate Zed editor with Vikat to use any AI provider for code assistance with virtual keys and observability.
- [Roo Code](https://docs.vikat.ai/cli-agents/roo-code.md): Use Roo Code with Vikat to access any AI provider through a powerful VS Code extension with virtual keys and observability.
- [Open WebUI](https://docs.vikat.ai/cli-agents/open-webui.md): Integrate Open WebUI with Vikat to access any AI provider through a modern open-source chat interface with virtual keys and observability.

### SDKs & Frameworks

- [LiteLLM SDK](https://docs.vikat.ai/integrations/litellm-sdk.md): Use Vikat as a drop-in proxy for LiteLLM applications with zero code changes.
- [Langchain SDK](https://docs.vikat.ai/integrations/langchain-sdk.md): Use Vikat as a drop-in proxy for Langchain applications with zero code changes.
- [Pydantic AI SDK](https://docs.vikat.ai/integrations/pydanticai-sdk.md): Use Vikat as a drop-in proxy for Pydantic AI agents with zero code changes.
- [Passthrough](https://docs.vikat.ai/integrations/passthrough.md): Forward provider-native requests through Vikat with full core pipeline processing, including logs and observability.

#### OpenAI SDK

- [Overview](https://docs.vikat.ai/integrations/openai-sdk/overview.md): Use Vikat as a drop-in replacement for OpenAI API with full compatibility and enhanced features.
- [Files and Batch API](https://docs.vikat.ai/integrations/openai-sdk/files-and-batch.md): Upload files and create batch jobs for asynchronous processing using the OpenAI SDK through Vikat across multiple providers.

#### Anthropic SDK

- [Overview](https://docs.vikat.ai/integrations/anthropic-sdk/overview.md): Use Vikat as a drop-in replacement for Anthropic API with full compatibility and enhanced features.
- [Files and Batch API](https://docs.vikat.ai/integrations/anthropic-sdk/files-and-batch.md): Upload files and create batch jobs for asynchronous processing using the Anthropic SDK through Vikat across multiple providers.

#### Bedrock SDK

- [Overview](https://docs.vikat.ai/integrations/bedrock-sdk/overview.md): Use Vikat as a Bedrock-compatible gateway for the Converse and Invoke APIs, with Vikat features on top.
- [Files and Batch API](https://docs.vikat.ai/integrations/bedrock-sdk/files-and-batch.md): Manage S3-based files and batch inference jobs using the AWS Bedrock SDK (boto3) through Vikat across multiple providers.

#### GenAI SDK

- [Overview](https://docs.vikat.ai/integrations/genai-sdk/overview.md): Use Vikat as a drop-in replacement for Google GenAI API with full compatibility and enhanced features.

### Identity Providers (SSO)

#### Okta

- [SSO using OIDC](https://docs.vikat.ai/enterprise/setting-up-okta/oidc.md): Configure Okta as your identity provider for Vikat Enterprise using OpenID Connect.
- [Setup SCIM](https://docs.vikat.ai/enterprise/setting-up-okta/scim.md): Enable real-time user and group provisioning from Okta to Vikat Enterprise using SCIM 2.0.

#### Microsoft Entra

- [SSO using OIDC](https://docs.vikat.ai/enterprise/setting-up-entra/oidc.md): Configure Microsoft Entra ID (Azure AD) as your identity provider for Vikat Enterprise using OpenID Connect.

#### Zitadel

- [SSO using OIDC](https://docs.vikat.ai/enterprise/setting-up-zitadel/oidc.md): Configure Zitadel (cloud or self-hosted) as your identity provider for Vikat Enterprise using OpenID Connect.

#### Keycloak

- [SSO using OIDC](https://docs.vikat.ai/enterprise/setting-up-keycloak/oidc.md): Configure Keycloak as your identity provider for Vikat Enterprise using OpenID Connect.

#### Google Workspace

- [Setting up Google Workspace](https://docs.vikat.ai/enterprise/setting-up-google-workspace.md): Step-by-step guide to configure Google Workspace as your identity provider for Vikat Enterprise SSO and Directory-based user provisioning.

#### Auth0

- [SSO using OIDC](https://docs.vikat.ai/enterprise/setting-up-auth0/oidc.md): Configure Auth0 as your identity provider for Vikat Enterprise using OpenID Connect.

#### Generic OIDC

- [SSO using OIDC](https://docs.vikat.ai/enterprise/setting-up-generic-oidc/oidc.md): Configure any standard OpenID Connect provider as your identity provider for Vikat Enterprise.
- [Setup SCIM](https://docs.vikat.ai/enterprise/setting-up-generic-oidc/scim.md): Enable real-time user and group provisioning from any SCIM 2.0-capable identity provider to Vikat Enterprise.

### Content Safety (Guardrails)

- [Secrets Detection](https://docs.vikat.ai/enterprise/guardrails/secrets-detection.md): Catch leaked API keys, tokens and private keys in prompts and responses using regex guardrails.
- [Custom Regex](https://docs.vikat.ai/enterprise/guardrails/custom-regex.md): Write organisation-specific content rules with RE2 regular expressions, and use the built-in PII patterns.

### Observability

- [Built-in Observability](https://docs.vikat.ai/features/observability/default.md): Monitor and analyze every AI request and response in real-time. Track performance, debug issues, and gain insights into your AI application's behavior with comprehensive request tracing.
- [OpenTelemetry (OTel)](https://docs.vikat.ai/features/observability/otel.md): Integrate with OpenTelemetry collectors for enterprise observability and distributed tracing
- [Prometheus](https://docs.vikat.ai/features/observability/prometheus.md): Monitor Vikat metrics with Prometheus scraping or Push Gateway for multi-node deployments
- [Kafka](https://docs.vikat.ai/features/observability/kafka.md): Stream Vikat request traces as JSON to a Kafka topic for custom analytics, archival, and downstream processing

### Vector Databases

- [Weaviate](https://docs.vikat.ai/integrations/vector-databases/weaviate.md): Weaviate vector database integration for semantic caching in Vikat.
- [Redis / Valkey](https://docs.vikat.ai/integrations/vector-databases/redis.md): Redis and Valkey vector store integration for semantic caching in Vikat.
- [Qdrant](https://docs.vikat.ai/integrations/vector-databases/qdrant.md): Qdrant vector database integration for semantic caching in Vikat.
- [Pinecone](https://docs.vikat.ai/integrations/vector-databases/pinecone.md): Pinecone vector database integration for semantic caching in Vikat.

## Deployment Guides

### Platform specific guides

- [Terraform + k8s](https://docs.vikat.ai/deployment-guides/k8s.md): Deploy Vikat as a service in Kubernetes clusters across AWS, Azure, and GCP using Terraform
- [ECS](https://docs.vikat.ai/deployment-guides/ecs.md): Deploy Vikat as a service in ECS AWS clusters
- [fly.io](https://docs.vikat.ai/deployment-guides/fly.md): This guide explains how to deploy Vikat on fly.io

### Config as Code

#### Helm

- [Quick Start](https://docs.vikat.ai/deployment-guides/helm.md): Deploy Vikat on Kubernetes using the official Helm chart - quickstart for OSS and Enterprise
- [Values Reference](https://docs.vikat.ai/deployment-guides/helm/values.md): Complete reference for Vikat Helm chart values - key parameters, how to supply them, and links to example files
- [Client Configuration](https://docs.vikat.ai/deployment-guides/helm/client.md): Configure the Vikat client: connection pool, logging, CORS, header filtering, compat shims, and MCP settings
- [Provider Setup](https://docs.vikat.ai/deployment-guides/helm/providers.md): Configure LLM providers in the Vikat Helm chart - API keys, cloud-native auth, and self-hosted endpoints
- [Storage](https://docs.vikat.ai/deployment-guides/helm/storage.md): Configure Vikat storage backends in Helm - SQLite, PostgreSQL (embedded and external), per-store overrides, and S3/GCS object storage for logs
- [Plugins](https://docs.vikat.ai/deployment-guides/helm/plugins.md): Configure Vikat plugins in Helm - telemetry, logging, semantic cache, OpenTelemetry, Datadog, governance, and custom plugins
- [Governance](https://docs.vikat.ai/deployment-guides/helm/governance.md): Configure Vikat governance in Helm - budgets, rate limits, virtual keys, routing rules, and admin authentication
- [Guardrails](https://docs.vikat.ai/deployment-guides/helm/guardrails.md): Configure guardrails providers and rules in Vikat Helm deployments
- [Secret Management](https://docs.vikat.ai/deployment-guides/helm/secret-management.md): Configure AWS Secrets Manager, GCP Secret Manager, or HashiCorp Vault in Vikat Helm deployments
- [Cluster Mode & HA](https://docs.vikat.ai/deployment-guides/helm/cluster.md): Run Vikat in a multi-replica cluster with gossip-based peer discovery, distributed state sync, and high-availability configuration
- [Troubleshooting](https://docs.vikat.ai/deployment-guides/helm/troubleshooting.md): Diagnose and fix common issues with Vikat Helm deployments - pods, database, ingress, secrets, PVCs, and performance

#### config.json

- [Quick Start](https://docs.vikat.ai/deployment-guides/config-json.md): Configure Vikat using a config.json file - GitOps-friendly, no-UI deployments, and multinode OSS setups
- [Source of Truth & Reconciliation](https://docs.vikat.ai/deployment-guides/config-json/source-of-truth.md): How config.json, the config store, split mode, and authoritative file sync interact at startup
- [Schema Reference](https://docs.vikat.ai/deployment-guides/config-json/schema-reference.md): All top-level keys available in config.json, their types, and where each is documented
- [Client Configuration](https://docs.vikat.ai/deployment-guides/config-json/client.md): Configure the Vikat client in config.json - connection pool, logging, CORS, header filtering, compat shims, and MCP settings
- [Provider Setup](https://docs.vikat.ai/deployment-guides/config-json/providers.md): Configure LLM providers in config.json - API keys, cloud-native auth, per-provider network settings, and self-hosted endpoints
- [Storage](https://docs.vikat.ai/deployment-guides/config-json/storage.md): Configure Vikat storage backends in config.json - config_store, logs_store, vector_store, and object storage for logs
- [Plugins](https://docs.vikat.ai/deployment-guides/config-json/plugins.md): Configure Vikat plugins in config.json - semantic cache, OpenTelemetry, Datadog, and custom plugins
- [Governance](https://docs.vikat.ai/deployment-guides/config-json/governance.md): Seed virtual keys, budgets, rate limits, routing rules, and admin auth in config.json
- [Cluster](https://docs.vikat.ai/deployment-guides/config-json/cluster.md): Configure enterprise cluster mode in config.json using peers or automatic discovery
- [Guardrails](https://docs.vikat.ai/deployment-guides/config-json/guardrails.md): Configure the content policy declaratively in config.json, through the plugins array.
- [Secret Management](https://docs.vikat.ai/deployment-guides/config-json/secret-management.md): Configure AWS Secrets Manager, GCP Secret Manager, or HashiCorp Vault in config.json using config_store.vault_store

### Enterprise Deployment

- [Overview](https://docs.vikat.ai/deployment-guides/enterprise/overview.md): Deploy Vikat Enterprise in your cloud environment with secure, private container image distribution
- [AWS Deployment](https://docs.vikat.ai/deployment-guides/enterprise/aws.md): Deploy Vikat Enterprise on AWS using ECR with IRSA or IAM Task Roles
- [GCP Deployment](https://docs.vikat.ai/deployment-guides/enterprise/gcp.md): Deploy Vikat Enterprise on GCP using Artifact Registry with Workload Identity
- [Azure Deployment](https://docs.vikat.ai/deployment-guides/enterprise/azure.md): Deploy Vikat Enterprise on Azure AKS using Workload Identity Federation to GCP Artifact Registry
- [On-Premise Deployment](https://docs.vikat.ai/deployment-guides/enterprise/on-premise.md): Deploy Vikat Enterprise in on-premise or air-gapped environments using Docker credentials

### Common setup instructions

- [Install make command](https://docs.vikat.ai/deployment-guides/how-to/install-make.md): This guide explains how to install make command.
- [Multinode Deployment](https://docs.vikat.ai/deployment-guides/how-to/multinode.md): Deploy multiple Vikat nodes with shared configuration for high availability in OSS deployments
- [Nginx reverse proxy](https://docs.vikat.ai/deployment-guides/how-to/nginx-reverse-proxy.md): Run Vikat behind NGINX with streaming-safe settings for SSE and WebSocket traffic
- [Security best practices](https://docs.vikat.ai/deployment-guides/how-to/security-best-practices.md): Best practices for hosting Vikat on the public internet: strong dashboard credentials, enforced inference auth, locked-down CORS, and reverse-proxy security headers.
- [Air-Gapped Deployment](https://docs.vikat.ai/deployment-guides/how-to/airgapped.md): Run Vikat in environments without outbound internet access.
- [Docker Performance Tuning](https://docs.vikat.ai/deployment-guides/docker-tuning.md): Optimize Vikat container performance with Go runtime tuning, resource limits, and system configuration

- [API (403 pages)](https://docs.vikat.ai/_llms/api.md): Documentation for API.

## Contributing

### Contributing

- [Setting up the repository](https://docs.vikat.ai/contributing/setting-up-repo.md): Complete guide to setting up the Vikat repository for local development.
- [Raising a Pull Request](https://docs.vikat.ai/contributing/raising-a-pr.md): Guidelines for submitting high-quality pull requests to Vikat.
- [Code Conventions](https://docs.vikat.ai/contributing/code-conventions.md): Code style and convention guidelines for contributing to Vikat.
- [Adding a new provider](https://docs.vikat.ai/contributing/adding-a-provider.md): Learn how to contribute a new provider to Vikat.
- [Adding config store](https://docs.vikat.ai/contributing/adding-a-configstore.md): Learn how to contribute a backend for the config store in Vikat
- [Adding a log store](https://docs.vikat.ai/contributing/adding-a-logstore.md): Learn how to contribute a backend for the log store in Vikat
- [Adding a vector store](https://docs.vikat.ai/contributing/adding-a-vectorstore.md): Learn how to contribute a backend for the vector store in Vikat

### Core Architecture

- [Concurrency](https://docs.vikat.ai/architecture/core/concurrency.md): Deep dive into Vikat's advanced concurrency architecture - worker pools, goroutine management, channel-based communication, and resource isolation patterns.
- [Request Flow](https://docs.vikat.ai/architecture/core/request-flow.md): Deep dive into Vikat's request processing pipeline - from transport layer ingestion through provider execution to response delivery.
- [Model Context Protocol (MCP)](https://docs.vikat.ai/architecture/core/mcp.md): Deep dive into Vikat's Model Context Protocol (MCP) integration - how external tool discovery, execution, and integration work internally.
- [Plugins](https://docs.vikat.ai/architecture/core/plugins.md): Deep dive into Vikat's extensible plugin architecture - how plugins work internally, lifecycle management, execution model, and integration patterns.

### Framework

- [What is framework?](https://docs.vikat.ai/architecture/framework/what-is-framework.md): Framework is Vikat's shared storage and utilities SDK package that provides common database interfaces and logic for the plugin ecosystem.
- [Model Catalog](https://docs.vikat.ai/architecture/framework/model-catalog.md): A centralized system for managing model information, pricing, and capabilities across all supported AI providers.
- [Config Store](https://docs.vikat.ai/architecture/framework/config-store.md): A persistent and flexible configuration management system for Vikat, supporting multiple database backends.
- [Log Store](https://docs.vikat.ai/architecture/framework/log-store.md): A robust and queryable system for persisting API request and response logs, with support for multiple database backends.
- [Vector Store](https://docs.vikat.ai/architecture/framework/vector-store.md): Vector database implementations for semantic search, embeddings storage, and AI-powered features in Vikat.
- [Streaming](https://docs.vikat.ai/architecture/framework/streaming.md): Framework utility for aggregating and processing real-time stream chunks from AI providers

## Benchmarks

- [Getting Started](https://docs.vikat.ai/benchmarking/getting-started.md): Introduction to Vikat's performance capabilities and how to choose the right instance size for your workload.
- [t3.medium](https://docs.vikat.ai/benchmarking/t3.medium.md): Detailed performance metrics and analysis for Vikat running on AWS t3.medium instances (2 vCPUs, 4GB RAM).
- [t3.xlarge](https://docs.vikat.ai/benchmarking/t3.xl.md): Detailed performance metrics and analysis for Vikat running on AWS t3.xlarge instances (4 vCPUs, 16GB RAM).
- [Run Your Own Benchmarks](https://docs.vikat.ai/benchmarking/run-your-own-benchmarks.md): Step-by-step guide to benchmark Vikat in your own environment using the official benchmarking tool.

## OpenAPI Specs

- [openapi](/openapi/openapi.json)

> The links below point to documentation indexes. Follow each `/_llms/` index recursively until you reach documentation pages.

## Indexes

- [OSS (103 pages)](https://docs.vikat.ai/_llms/oss.md): Documentation for OSS.
- [API (403 pages)](https://docs.vikat.ai/_llms/api.md): Documentation for API.