Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
File size: 2,729 Bytes
94753b6 ca548dc 94753b6 ca548dc 94753b6 9abd9af 94753b6 9abd9af 94753b6 9abd9af 94753b6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 |
import type { InferenceTask, Options, RequestArgs } from "../types";
import { HF_HUB_URL } from "./getDefaultTask";
import { isUrl } from "./isUrl";
const HF_INFERENCE_API_BASE_URL = "https://api-inference.huggingface.co";
/**
* Loaded from huggingface.co/api/tasks if needed
*/
let tasks: Record<string, { models: { id: string }[] }> | null = null;
/**
* Helper that prepares request arguments
*/
export async function makeRequestOptions(
args: RequestArgs & {
data?: Blob | ArrayBuffer;
stream?: boolean;
},
options?: Options & {
/** When a model can be used for multiple tasks, and we want to run a non-default task */
forceTask?: string | InferenceTask;
/** To load default model if needed */
taskHint?: InferenceTask;
}
): Promise<{ url: string; info: RequestInit }> {
// eslint-disable-next-line @typescript-eslint/no-unused-vars
const { accessToken, model: _model, ...otherArgs } = args;
let { model } = args;
const {
forceTask: task,
includeCredentials,
taskHint,
wait_for_model,
use_cache,
dont_load_model,
...otherOptions
} = options ?? {};
const headers: Record<string, string> = {};
if (accessToken) {
headers["Authorization"] = `Bearer ${accessToken}`;
}
if (!model && !tasks && taskHint) {
const res = await fetch(`${HF_HUB_URL}/api/tasks`);
if (res.ok) {
tasks = await res.json();
}
}
if (!model && tasks && taskHint) {
const taskInfo = tasks[taskHint];
if (taskInfo) {
model = taskInfo.models[0].id;
}
}
if (!model) {
throw new Error("No model provided, and no default model found for this task");
}
const binary = "data" in args && !!args.data;
if (!binary) {
headers["Content-Type"] = "application/json";
}
if (wait_for_model) {
headers["X-Wait-For-Model"] = "true";
}
if (use_cache === false) {
headers["X-Use-Cache"] = "false";
}
if (dont_load_model) {
headers["X-Load-Model"] = "0";
}
const url = (() => {
if (isUrl(model)) {
return model;
}
if (task) {
return `${HF_INFERENCE_API_BASE_URL}/pipeline/${task}/${model}`;
}
return `${HF_INFERENCE_API_BASE_URL}/models/${model}`;
})();
/**
* For edge runtimes, leave 'credentials' undefined, otherwise cloudflare workers will error
*/
let credentials: RequestCredentials | undefined;
if (typeof includeCredentials === "string") {
credentials = includeCredentials as RequestCredentials;
} else if (includeCredentials === true) {
credentials = "include";
}
const info: RequestInit = {
headers,
method: "POST",
body: binary
? args.data
: JSON.stringify({
...otherArgs,
options: options && otherOptions,
}),
...(credentials && { credentials }),
signal: options?.signal,
};
return { url, info };
}
|