32 lines
1.1 KiB
JavaScript
32 lines
1.1 KiB
JavaScript
// ============================================================
|
||
// LLM 调用工具 — 独立于 agent-worker，直接 fetch API
|
||
// ============================================================
|
||
|
||
const BASE_URL = (process.env.LLM_BASE_URL || process.env.OPENAI_BASE_URL || "https://api.openai.com/v1").replace(/\/+$/, "");
|
||
const MODEL = process.env.LLM_MODEL || process.env.OPENAI_MODEL || "gpt-4o";
|
||
const API_KEY = process.env.LLM_API_KEY || process.env.OPENAI_API_KEY || "";
|
||
|
||
async function callLLM({ messages, model, maxTokens, temperature }) {
|
||
const url = BASE_URL + "/chat/completions";
|
||
const body = {
|
||
model: model || MODEL,
|
||
messages: messages,
|
||
max_tokens: maxTokens || 2048,
|
||
temperature: temperature ?? 0.7
|
||
};
|
||
const resp = await fetch(url, {
|
||
method: "POST",
|
||
headers: {
|
||
"Content-Type": "application/json",
|
||
"Authorization": "Bearer " + API_KEY
|
||
},
|
||
body: JSON.stringify(body)
|
||
});
|
||
if (!resp.ok) {
|
||
const errText = await resp.text().catch(() => "");
|
||
throw new Error("LLM " + resp.status + ": " + errText.slice(0, 200));
|
||
}
|
||
return await resp.json();
|
||
}
|
||
|
||
module.exports = { callLLM }; |