perf: customizable embedding chunk size via env var (#5494)

* perf: customizable embedding chunk size via env var

* Update .env.template

---------

Co-authored-by: Archer <545436317@qq.com>

packages/service/core/ai/embedding/index.ts

@@ -24,7 +24,7 @@ export async function getVectorsByText({ model, input, type, headers }: GetVecto
   const formatInput = Array.isArray(input) ? input : [input];
   const formatInput = Array.isArray(input) ? input : [input];
 
 
   // 20 size every request
   // 20 size every request
-  const chunkSize = 20;
+  const chunkSize = parseInt(process.env.EMBEDDING_CHUNK_SIZE || '10');
   const chunks = [];
   const chunks = [];
   for (let i = 0; i < formatInput.length; i += chunkSize) {
   for (let i = 0; i < formatInput.length; i += chunkSize) {
     chunks.push(formatInput.slice(i, i + chunkSize));
     chunks.push(formatInput.slice(i, i + chunkSize));

projects/app/.env.template

@@ -13,6 +13,9 @@ ROOT_KEY=fdafasd
 # 强制将图片转成 base64 传递给模型
 # 强制将图片转成 base64 传递给模型
 MULTIPLE_DATA_TO_BASE64=true
 MULTIPLE_DATA_TO_BASE64=true
 
 
+# 临时解决-emb 模型单次并发量
+EMBEDDING_CHUNK_SIZE=10
+
 # 是否隐藏版权信息配置，只有值为 'true' 时隐藏
 # 是否隐藏版权信息配置，只有值为 'true' 时隐藏
 HIDE_CHAT_COPYRIGHT_SETTING=
 HIDE_CHAT_COPYRIGHT_SETTING=
 
 
@@ -96,3 +99,4 @@ CONFIG_JSON_PATH=
 SIGNOZ_BASE_URL=
 SIGNOZ_BASE_URL=
 SIGNOZ_SERVICE_NAME=
 SIGNOZ_SERVICE_NAME=
 SIGNOZ_STORE_LEVEL=warn
 SIGNOZ_STORE_LEVEL=warn
+