Skip to content

Testmiljø

Hurtig, gennemsigtig adgang til flere AI-modeller.

Servertid 2026-10-17 02:07:51 UTC
Demo — not your personal account

The balance and activity are examples, not your funds or requests. This preview cannot send AI requests or make payments. Your real sign-in is separate.

Go to my account How to connect AI
Demo preview

Testmiljø

Preview the interactive request builder, protocol controls, response artifacts and accounting diagnostics. No request can be sent from Demo mode.

View onlyNo session token is created and no API endpoint is called.
ProtokolOpenAI Responses
API-endepunkt/v1/responses
TransportStreaming
HTTP-status200
Duration1.42 s
Request IDDEMO-REQUEST
Summarize why response caching can reduce LLM API cost.
Response caching can avoid recomputing repeated context and lets supported providers bill reused prompt material at a lower cache-read rate. The exact savings depend on the model and cache policy.

Usage & diagnostics

Input842 tokens
Output96 tokens
Cache read12,480 tokens
Response cost$0.0032
Raw request
{
  "model": "claude-sonnet-5",
  "input": "Summarize why response caching can reduce LLM API cost.",
  "stream": true
}
Raw response / events
{"type":"response.completed","status":"completed","request_id":"DEMO-REQUEST"}