On this page

POST /v1/sellers/{id}/asks

Rest an ask on the model book. Rejected while the seller is Suspended or Banned.

Auth: X-Seller-API-Key.

See rate limits and errors. Capacity sizing and price derivation are in the seller flow.

Request

field	type	notes
`model_id`	string	Must be in `supported_models`
`price`	decimal string	Per 1,000 tokens, strictly positive
`capacity`	number	Tokens offered
`ttl_seconds`	number	Defaults to 300

Response

ask_id, seller_id, model_id, price, capacity, remaining, created_at, expires_at.

Errors

unauthorized, forbidden, invalid_model, not_found. Status codes on Errors.

cURL

curl -s http://127.0.0.1:8080/v1/sellers/SELLER_ID/asks \
  -H "X-Seller-API-Key: sk-seller" \
  -H "Content-Type: application/json" \
  -d '{"model_id":"gpt-4","price":"10","capacity":100,"ttl_seconds":300}'

Rust

use token_gateway_sdk::{Decimal, RegisterSellerRequest, SellerClient, SellerKey, SubmitAskRequest};
// {"model_id":"gpt-4","price":"10","capacity":100,"ttl_seconds":300}

let mut client = SellerClient::new("http://127.0.0.1:8080", None);
let registered = client
    .register(&RegisterSellerRequest {
        name: "acme".into(),
        endpoint_url: "http://127.0.0.1:9000/infer".into(),
        supported_models: vec!["gpt-4".into()],
        api_key: "sk-seller".into(),
    })
    .await?;
client.set_key(SellerKey(registered.api_key.clone()));
let ask = client
    .submit_ask(
        registered.seller_id,
        &SubmitAskRequest {
            model_id: "gpt-4".into(),
            price: Decimal::from(10),
            capacity: 100,
            ttl_seconds: Some(300),
        },
    )
    .await?;
let _ = ask.ask_id;

Python

# {"model_id":"gpt-4","price":"10","capacity":100,"ttl_seconds":300}
from token_gateway import SellerTokenGateway

gw = SellerTokenGateway("http://127.0.0.1:8080")
seller = gw.register_seller(name="acme", endpoint_url="http://127.0.0.1:9000/infer", supported_models=["gpt-4"], api_key="sk-seller")
gw.submit_ask(seller.seller_id, model_id="gpt-4", price="10", capacity=100, ttl_seconds=300)

TypeScript

// {"model_id":"gpt-4","price":"10","capacity":100,"ttl_seconds":300}
import { SellerClient } from "@ai-token-gateway/sdk";

const seller = new SellerClient({ baseUrl: "http://127.0.0.1:8080" });
const registered = await seller.register({
  name: "acme", endpoint_url: "http://127.0.0.1:9000/infer",
  supported_models: ["gpt-4"], api_key: "sk-seller",
});
seller.setApiKey(registered.api_key);
await seller.submitAsk(registered.seller_id, { model_id: "gpt-4", price: "10", capacity: 100, ttl_seconds: 300 });