NaraRouter

NaraRouter - Unified AI Gateway

NaraRouter is a production-ready AI infrastructure platform that simplifies access to multiple AI models through a single, stable OpenAI-compatible API endpoint. It acts as a unified gateway routing requests across 15+ leading AI providers including OpenAI, Anthropic, Google, DeepSeek, Meta, Mistral, and more.

Key Features

• Single API Endpoint: One base URL (https://router.bynara.id/v1) for OpenAI-compatible /v1/chat/completions and Anthropic-compatible /v1/messages endpoints
• Multi-Provider Access: Route requests across Anthropic, Google AI, OpenAI, DeepSeek, Zhipu AI, Moonshot AI, Mistral, Tencent, Meta, and others
• Usage Analytics & Tracking: Real-time monitoring of requests, tokens, latency, success rates, and credit consumption
• Credit & Voucher Management: Prepaid credits, vouchers, and usage limits for predictable costs
• Rate Limiting & Controls: Per-user, per-API key, per-provider, or per-model limits
• Request Logs: Debug with status logs, latency, token usage, retries, and error details
• BYOK (Bring Your Own Key): Connect your own provider keys while keeping unified tracking
• Free Tier: Free models included with no credit card required
• Developer Tool Integration: Works with NaraCLI, Claude Code, Opencode, Factory CLI, VS Code, and more

Pricing Plans

Offers flexible plans including Free (Rp 0/day, 7M tokens/day), Freemium (Rp 5K/day), Mimo Lite/Plus/Pro, DeepSeek Lite/Plus/Pro, Minimax, and GPT Families plans with daily token caps ranging from 7M to 35M tokens/day and rate limits from 10-60 req/min.

Use Cases

• AI Application Development: Simplify integration by using one SDK/client for multiple models
• Cost Control: Track and limit AI spending across teams and projects
• Model Experimentation: Easily switch between providers without code changes
• Production Deployments: Enterprise-grade infrastructure with reliability and monitoring
• Coding Agents: Direct integration with Claude Code, Opencode, and other AI coding tools

Technical Details

• OpenAI SDK compatible (change only base URL and API key)
• Anthropic SDK compatible
• Streaming support
• Reasoning models support with reasoning_effort parameter
• Image and video generation endpoints
• Webhook and async processing for video generation