AivoryLabs.in
Initializing0%
Applied AI Engineering / Bengaluru
Chat with usBook a call
All Case StudiesOWN PRODUCT · AI GATEWAY

Axora

The open-source AI gateway that runs on your servers — one endpoint, five providers, automatic failover, zero lock-in.

Visit websiteLive DemoAPI DocsGitHub
FastAPINext.js 16TypeScriptPostgreSQL 16Redis 7SSE StreamingRAGDocker
Axora homepage hero
Screenshots
Axora screenshot 1
Axora screenshot 2
Axora screenshot 3
Axora screenshot 4
Axora screenshot 5
Axora screenshot 6
Axora screenshot 7
Axora screenshot 8
5Providers
One /v1/chatEndpoint
LiveStatus
THE CHALLENGE

Why it mattered

Teams working with multiple AI models end up with fragmented tooling: separate keys, separate interfaces, and no way to route a request to the best model for the job. When a provider degrades, requests just fail. Axora is our answer to that mess.

OUR SOLUTION

What we built

We designed Axora as a self-hosted AI gateway that sits between your application and every major LLM provider. Send one request to /v1/chat and Axora routes it across Groq, OpenRouter, Gemini, OpenAI, and Claude with automatic failover — if one provider goes down, the next picks up. It ships with a ChatGPT-style web interface, streaming SSE responses, provider-scoped API keys, per-key usage analytics, an admin dashboard, and a built-in RAG knowledge base. It's fully self-hosted: your data never leaves your servers.

Key Outcome

A production-ready, MIT-licensed open-source AI gateway — one endpoint, five providers, automatic failover, and zero vendor lock-in. Live at axora.aivorylabs.in.

WHAT WE BUILT

Features & main functionality

The core functionality and features we built for Axora.

01

Multi-Provider Routing

Groq, OpenRouter, Gemini, OpenAI, and Claude behind one endpoint — with automatic failover when a provider goes down.

02

Streaming Responses

Real-time SSE streaming with token-by-token display, across every connected provider.

03

Chat Interface

A polished ChatGPT-style UI with model switching, conversation history, and auto-generated titles.

04

Message Actions

Copy and edit messages inline — editing auto-resends and streams a fresh response.

05

API Key Management

Provider-scoped keys with SHA-256 hashing, per-key usage tracking, and instant revocation.

06

Usage Analytics

Track requests, tokens, and cost per provider, per model, and per user from a live dashboard.

07

Knowledge Base (RAG)

Upload markdown docs, auto-chunk, embed, and retrieve relevant context into every conversation.

08

Admin Dashboard

System health monitoring, user management, and error tracking for the whole deployment.

← Previous ProjectHotel AmaravathiNext Project →HMS
CONTACT

Want something like Axora?

We design and ship fast, reliable software for businesses that need results.

Contact us