gpt-4.1
openai/gpt-4.1Available
Openai-r openai/gpt-4.1 - chat model
Retail priceinput $4e-06 · output $1.6e-05 · cached input $1e-06
Provideropenai
Capabilitychat
InterfacePOST /v1/run
Request example
Every paid call needs a unique Idempotency-Key. The same key and payload replay the original result without a second charge.
curl https://router.tycoon.cool/v1/run \
+ -H "Authorization: Bearer $TYCOONROUTER_API_KEY" \
+ -H "Content-Type: application/json" \
+ -H "Idempotency-Key: $(uuidgen)" \
+ -d '{"model":"openai/gpt-4.1","inputs":{"messages":[{"role":"user","content":"Hello"}]}}'Response contract
Synchronous routes return the provider result with TycoonRouter request metadata. Async routes return a request ID plus status and cancel URLs.
| Header | Purpose |
|---|---|
| x-request-id | Immutable Router request identity |
| x-idempotent-replay | True when the response is a replay |
| x-tycoonrouter-charge-micros | Settled retail charge when available |
Production safeguards
Unknown prices, unbounded requests and uncertain provider outcomes fail closed before a second paid dispatch. Usage and wallet events remain queryable from the account page.