Built for developers worldwide

SeedRouter API Reliable access Fast setup

Selected GPT models: 10% of official pricing. Claude, Grok and Gemini: 30%. Use the desktop setup tool for Codex CLI and Claude Code.

Register Get setup tool

Connect SeedRouter to Codex

Use our exclusive configuration approach while keeping official account sign-in and all native features.

Integration guide 

            


          

        

      


      

        

          

            
Smart routing to stable channels

            
SeedRouter automatically monitors channel stability and does everything possible to keep the service running normally.

          

          

            

              

                

                  

                    Current multi-route model
                    Healthy pool
                  

                  active
                

                

                  

                    

                      Model route Aprimary pool
                    

                    

                    

                      135ms
                    

                    
ok

                  

                  

                    

                      Model route Breserve pool
                    

                    

                    

                      175ms
                    

                    
ok

                  

                  

                    

                      Model route Ccompatible route
                    

                    

                    

                      215ms
                    

                    
ok

                  

                  

                    

                      Model route Dfallback route
                    

                    

                    

                      255ms
                    

                    
ok

                  

                

                

                  

                    selected
                    Anthropic · US-east
                  

                  

                    Lowest latency across 4 routes.
                  

                

              

            

          

        

      


      

      

        

          

            
Access the latest models at ultra-low prices

            
Usage-based pricing for each model. See the latest price table for current rates.

            
Image generation and editing: for gpt-image-2.5 models and the gpt-image-2 alias, each successful call costs the higher of token usage or $0.02, not both. Failed calls are free; internal retries are not billed again. Existing free trials are unchanged.

          

          

            

              

                

                  

                    Cost comparison
                    Calculate how much you can save
                  

                  
                

                

                  

                    SeedRouter cost
                    $-
                    for the selected monthly volume
                  

                  

                    Official cost
                    $-
                    for the same usage
                  

                  

                    Estimated savings
                    $-
                    compared with the official API
                  

                

                

                  
                  
                  
                

              

            

          

        

      


      

        

          

            
Works with all kinds of Agent Harnesses.

            
Replace the Base URL and API key to connect the Agent Harness you already use.

          

          

            
            
            
            
            
            
            
            
            
          

        

      


      

      

        

          

            
Manage everything from one dashboard

            
No need to switch between providers.

          

          

            

              
              Usage overview
            

            

              
              API keys and models
            

            

              
              Request visibility
            

          

          

            
            

              

                Status
                Docs
                Feedback
              

              

                

                  
Welcome back, user.

                

                Refresh
              

              

                

                  Balance
                  $248.72
                  Wallet balance plus invite credit
                

                

                  Today spend
                  $18.43
                  Today's consumed quota
                

                

                  Requests
                  12,984
                  Total requests recorded
                

                

                  Models
                  Live
                  Available on this site
                

              

              

                

                  

                    Recent activity
                    Latest self logs
                  

                  

                    

                      2m ago
                      Flagship reasoning
                      $0.04
                    

                    

                      8m ago
                      General chat
                      $0.02
                    

                    

                      21m ago
                      Fast and lightweight
                      $0.01
                    

                  

                

                

                  

                    Allowed models
                    Site whitelist after filtering
                  

                  

                    Flagship reasoning
                    General chat
                    Fast and lightweight
                    Long context
                    Image generation
                    More models
                  

                

              

            

          

        

      


      

        

          

            
Before you switch

            
Common questions

            
SeedRouter is designed for teams that want model choice without turning provider operations into product work.

          

          

            

              Which is the official SeedRouter website?
              

                The only official SeedRouter website is seedrouter.net, launched in May 2026. Similarly named domains are not affiliated with SeedRouter, so register, top up and create API keys only on seedrouter.net.
                Official channels and impersonation notice
              

            

            

              What is SeedRouter?
              
SeedRouter is an AI model API gateway for developers: one OpenAI-compatible base URL and API key gives you GPT, Claude, Gemini, Grok and other models. It is built mainly for AI coding tools such as Codex CLI and Claude Code, bills by actual token usage, and fails over automatically when a model has multiple routes.

            

            

              How do I get an API key?
              
Register and create a key in the dashboard. New accounts receive two free popular-model trials, and keys can be scoped or revoked independently.

            

            

              How is pricing calculated?
              
Each model has its own input, output and cache rates. The homepage table comes from the production site configuration; the dashboard model catalog and usage logs are final.

            

            

              What happens if a provider goes down?
              
For models configured with multiple routes, SeedRouter monitors route health and can move a request to another available route without requiring a client change.

            

            

              Can I keep my current SDK?
              
Yes. OpenAI-compatible clients can usually migrate by changing the base URL and API key.

            

            

              Is my request data used for training?
              
No. SeedRouter passes requests through to the provider you're routed to and returns the response. We don't retain prompt or completion payloads beyond short-lived debugging logs you control.

            

            

              Still have a question?
              

                Need a human? Email us at
                [email protected]
                and we will reply within one business day.