Deploy Create Llm

//native

type Deepinfra = {

  token: string;

};

/**

 * Deploy Create Llm

 *

 */

export async function main(

  auth: Deepinfra,

  body: {

    model_name: string;

    gpu: "A100-80GB" | "H100-80GB" | "H200-141GB";

    num_gpus?: number;

    max_batch_size?: number;

    hf?: { repo: string; revision?: string; token?: string };

    settings?: { min_instances?: number; max_instances?: number };

  },

) {

  const url = new URL(`https://api.deepinfra.com/deploy/llm`);

  const response = await fetch(url, {

    method: "POST",

    headers: {

      "Content-Type": "application/json",

      Authorization: "Bearer " + auth.token,

    },

    body: JSON.stringify(body),

  });

  if (!response.ok) {

    const text = await response.text();

    throw new Error(`${response.status} ${text}`);

  }

  return await response.json();

}