All
Generative AI

LLM Router

Name a model and let Eden AI choose which provider serves it, by cost, latency or region, with automatic failover when a provider fails. Add fallbacks to switch to another model, or save the whole chain as a routing policy and call it by name.

LLM Router

Trusted by 500+ Companies, from Startups to Industry Leaders

TUTORIAL

How to route LLM requests with Eden AI?

LLM Router

What is an LLM Router?

An LLM router sits between your application and the model providers and decides where each request goes. Many open models are sold by several providers at different prices and speeds: the router picks the best one for each call, retries elsewhere when a provider fails, and can switch to a backup model. Your code keeps one endpoint and one model name.

Try now

Route by Cost or Region

Name the model, pick the objective. Add @eu for Europe.

curl https://api.edenai.run/v3/chat/completions \
  -H "Authorization: Bearer $EDENAI_API_KEY" \
  -H "Content-Type: application/json" -H "x-edenai-metadata: enabled" \
  -d '{"model": "kimi-k2.6", "routing": {"sort": "cost"}, "messages": [{"role": "user", "content": "Define AI gateway."}]}'
# → served=infomaniak/moonshotai/Kimi-K2.6 (cheapest of 8)
See documentation

237 Models, Several Providers

kimi-k2.6 alone has 8, from $0.67 (Infomaniak, CH) to $1.20 per 1M input tokens. Add up to 7 fallbacks, or save the chain as a routing policy and call it by name.

value

Value delivered

/ 1 /
save time

Save time

Integrate once and access hundreds of models without managing multiple APIs.

Model updates, provider changes, and new releases are handled transparently.

/ 2 /
Control costs

Reduce costs

Use the most efficient model for each need. Avoid vendor lock-in and adapt quickly to pricing or performance changes.

/ 3 /
Reduce risk

Reduce risk

Built-in fallback mechanisms protect applications from model outages. Routing flexibility allows rapid adaptation to evolving technologies and providers.

They are using LLM Router

No items found.
Solution

Got specific needs? We're here to help

AI API Gateway

5.5% platform fee

Access to hundreds of models with our unified API
Compare AI models accuracy, latency and price
Create multiple API keys for different projects
Cost and performance monitoring tools
Chat support (48h - working days)
Unlimited seats

Advanced AI Platform

Custom price

All the previous features +
High rate limits and bulk discounts
Private deployments / specific compliance needs
Custom latency / cost optimization
Dedicated support & SLA / Professional Services
Custom billing and invoicing structures

Other features

Batch API
Batch API
All
Generative AI

Send up to 5,000 chat completion requests in one JSONL file and collect the results when they are ready, at half the standard token price. The Eden AI Batch API follows the OpenAI Batch format and runs OpenAI and Anthropic models with the same API key.

MCP Server
MCP Server
All
Generative AI

Connect any MCP client or agent to the Eden AI hosted MCP server and give it 42 tools: OCR and document parsing, web search and scraping, speech, translation, image analysis and more. One URL, one API key, no server to run.

Custom Chatbot with RAG API
All
Generative AI

Unlock the power of AI with our Custom Chatbot featuring RAG. Seamlessly integrate advanced conversational capabilities into your applications with ease.

let’s start

Start building with Eden AI

A single interface to integrate the best AI technologies into your products.