LLM_Router

Agents buy router credits through AG3N7 STORE, then create a separate router API key for the OpenAI-compatible base URL. The router key is not the same credential as the purchase API key.

Agent_Flow

Example MCP-style transcript. Each next request uses IDs returned by the previous response.

text

1. catalog_search { category: "api_tokens", q: "router credits", limit: 1 }
2. catalog_checkout_requirements { catalogItemId }
3. quote_create {
     catalogItemId,
     quantity: 1,
     configuration: { creditAmountUsd: 25 }
   }
4. purchase_create { quoteId, idempotencyKey }
5. service_get { serviceId }
6. router_api_key_create { serviceId, label: "prod-router" }
7. GET https://ag3n7.store/openai/v1/models
8. POST https://ag3n7.store/openai/v1/chat/completions

Step_1_Request_Search

json

{
  "tool": "catalog_search",
  "input": {
    "category": "api_tokens",
    "q": "router credits",
    "limit": 1
  }
}

Step_1_Response

json

{
  "items": [
    {
      "id": "cat_router_credit_001",
      "sku": "llm-router-credit-balance",
      "name": "LLM Router Credits",
      "category": "api_tokens",
      "checkoutKind": "custom"
    }
  ]
}

Step_2_Request_Checkout_Requirements

json

{
  "tool": "catalog_checkout_requirements",
  "input": {
    "catalogItemId": "cat_router_credit_001"
  }
}

Step_2_Response

json

{
  "catalogItemId": "cat_router_credit_001",
  "checkoutKind": "custom",
  "serviceType": "llm_router",
  "requiredFields": [
    "creditAmountUsd"
  ],
  "managementCapabilities": [
    "router:api-key:create",
    "router:api-key:list",
    "router:usage:get"
  ]
}

Step_3_Request_Quote

json

{
  "tool": "quote_create",
  "input": {
    "catalogItemId": "cat_router_credit_001",
    "quantity": 1,
    "configuration": {
      "creditAmountUsd": 25
    }
  }
}

Step_3_Response

json

{
  "quote": {
    "id": "quote_router_001",
    "catalogItemId": "cat_router_credit_001",
    "totalPriceMinor": 2500,
    "currency": "usd",
    "displayName": "LLM Router Credits ($25)",
    "quoteDetails": {
      "creditAmountUsd": 25
    },
    "managementCapabilities": [
      "router:api-key:create",
      "router:usage:get"
    ]
  }
}

Step_4_Request_Purchase

json

{
  "tool": "purchase_create",
  "input": {
    "quoteId": "quote_router_001",
    "idempotencyKey": "router-credit-topup-001"
  }
}

Step_4_Response

json

{
  "purchase": {
    "id": "purchase_router_001",
    "status": "completed",
    "totalPriceMinor": 2500
  },
  "service": {
    "id": "svc_router_001",
    "serviceType": "llm_router",
    "displayName": "Primary Router Credit Pool"
  }
}

Step_5_Request_Service

json

{
  "tool": "service_get",
  "input": {
    "serviceId": "svc_router_001"
  }
}

Step_5_Response

json

{
  "service": {
    "id": "svc_router_001",
    "status": "active",
    "serviceType": "llm_router",
    "displayName": "Primary Router Credit Pool",
    "router": {
      "creditBalanceUsd": "25.00",
      "baseUrl": "https://ag3n7.store/openai/v1",
      "supportedProviders": [
        "openai",
        "anthropic",
        "google",
        "kimi"
      ]
    }
  }
}

Step_6_Request_Router_Key

json

{
  "tool": "router_api_key_create",
  "input": {
    "serviceId": "svc_router_001",
    "label": "prod-router"
  }
}

Step_6_Response

json

{
  "apiKey": {
    "id": "router_key_001",
    "label": "prod-router",
    "prefix": "asr_live_",
    "secret": "asr_live_xxxxx.yyyyy",
    "baseUrl": "https://ag3n7.store/openai/v1"
  }
}

Step_7_Models_Response

json

{
  "object": "list",
  "data": [
    { "id": "gpt-5.4", "object": "model", "owned_by": "openai" },
    { "id": "claude-sonnet-4", "object": "model", "owned_by": "anthropic" },
    { "id": "gemini-2.5-pro", "object": "model", "owned_by": "google" },
    { "id": "kimi-k2.5", "object": "model", "owned_by": "kimi" }
  ]
}

Step_8_Request_Chat_Completions

bash

curl https://ag3n7.store/openai/v1/chat/completions \
  -H "Authorization: Bearer $AGENT_STORE_ROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4",
    "messages": [
      { "role": "system", "content": "Be concise." },
      { "role": "user", "content": "Summarize the deployment status." }
    ]
  }'

Step_8_Response

json

{
  "id": "chatcmpl_router_001",
  "object": "chat.completion",
  "model": "claude-sonnet-4",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Deployment is healthy. All regions are green and queue depth is normal."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 24,
    "completion_tokens": 20,
    "total_tokens": 44
  }
}

Compatible_Models

Current router catalog includes 14 listed models across OpenAI, Anthropic, Gemini, and Kimi.

OpenAI

GPT-5.4GPT-5.4 miniGPT-5.4 nano

Anthropic

Claude Opus 4.1Claude Sonnet 4Claude Haiku 4.5Claude Haiku 3.5

Google Gemini

Gemini 2.5 ProGemini 2.5 FlashGemini 2.5 Flash-LiteGemini 3.1 Flash-Lite Preview

Kimi

Kimi K2.5Kimi K2 0905Kimi K2 Thinking

Pricing_Basis

Provider rates are documented separately and currently shown with no markup.

GPT-5.4: input $2.5/1M • output $15/1M

GPT-5.4 mini: input $0.75/1M • output $4.5/1M

GPT-5.4 nano: input $0.2/1M • output $1.25/1M

Claude Opus 4.1: input $15/1M • output $75/1M

Open_Full_Pricing_Page