v4.6-4 (#473)

docSite/content/docs/installation/upgrading/46.md

@@ -50,5 +50,6 @@ curl --location --request POST 'https://{{host}}/api/admin/initv46-2' \
 1. 新增 - 团队空间
 2. 新增 - 多路向量（多个向量映射一组数据）
 3. 新增 - tts语音
-4. 线上环境新增 - ReRank向量召回，提高召回精度
-5. 优化 - 知识库导出，可直接触发流下载，无需等待转圈圈
+4. 新增 - 支持知识库配置文本预处理模型
+5. 线上环境新增 - ReRank向量召回，提高召回精度
+6. 优化 - 知识库导出，可直接触发流下载，无需等待转圈圈

docSite/content/docs/pricing.md

 ---
-title: '定价'
-description: 'FastGPT 的定价'
+title: '线上版定价'
+description: 'FastGPT 线上版定价'
 icon: 'currency_yen'
 draft: false
 toc: true
-weight: 10
+weight: 11
 ---
 
 ## Tokens 说明
@@ -15,7 +15,7 @@ weight: 10
 
 ## FastGPT 线上计费
 
-目前，FastGPT 线上计费也仅按 Tokens 使用数量为准。以下是详细的计费表（最新定价以线上表格为准，可在点击充值后实时获取）：
+使用: [https://fastgpt.run](https://fastgpt.run) 或  [https://ai.fastgpt.in](https://ai.fastgpt.in) 只需仅按 Tokens 使用数量扣费即可。可在 账号-使用记录 中查看具体使用情况，以下是详细的计费表（最新定价以线上表格为准，可在点击充值后实时获取）：
 
 {{< table "table-hover table-striped-columns" >}}
 | 计费项                 | 价格: 元/ 1K tokens（包含上下文） |

docSite/content/docs/use-cases/datasetEngine.md

 ---
 title: "知识库结构讲解"
-description: "本节会介绍 FastGPT 知识库结构设计，理解其 QA 的存储格式和检索格式，以便更好的构建知识库。这篇介绍主要以使用为主，详细原理不多介绍。"
+description: "本节会详细介绍 FastGPT 知识库结构设计，理解其 QA 的存储格式和多向量映射，以便更好的构建知识库。这篇介绍主要以使用为主，详细原理不多介绍。"
 icon: "dataset"
 draft: false
 toc: true
@@ -25,13 +25,21 @@ FastGPT 采用了 RAG 中的 Embedding 方案构建知识库，要使用好 Fast
 
 FastGPT 采用了 `PostgresSQL` 的 `PG Vector` 插件作为向量检索器，索引为`HNSW`。且`PostgresSQL`仅用于向量检索，`MongoDB`用于其他数据的存取。
 
-在`PostgresSQL`的表中，设置一个 `index` 字段用于存储向量、一个 `q` 字段用于存储向量对应的内容，以及一个 `a` 字段用于检索映射。之所以取字段为 `qa` 是由于一些历史缘故，无需完全解为 “问答对” 的格式。在实际使用过程中，可以利用`q`和`a`的组合，对检索后的内容做进一步的声明，提高大模型的理解力（注意，这里不直接提高搜索精度）。
+在`PostgresSQL`的表中，设置一个 `index` 字段用于存储向量，以及一个`data_id`用于在`MongoDB`中寻找对应的映射值。多个`index`可以对应一组`data_id`，也就是说，一组向量可以对应多组数据。在进行检索时，相同数据会进行合并。
 
-目前，提高向量搜索的精度，主要可以通过几种途径：
+![](/imgs/datasetSetting1.png)
 
-1. 精简`q`的内容，减少向量内容的长度：当`q`的内容更少，更准确时，检索精度自然会提高。但与此同时，会牺牲一定的检索范围，适合答案较为严格的场景。
-2. 更好分词分段：当一段话的结构和语义是完整的，并且是单一的，精度也会提高。因此，许多系统都会优化分词器，尽可能的保障每组数据的完整性。
-3. 多样性文本：为一段内容增加关键词、摘要、相似问题等描述性信息，可以使得该内容的向量具有更大的检索覆盖范围。
+## 多向量的目的和使用方式
+
+在一组数据中，如果我们希望它尽可能长，但语义又要在向量中尽可能提现，则没有办法通过一组向量来表示。因此，我们采用了多向量映射的方式，将一组数据映射到多组向量中，从而保障数据的完整性和语义的提现。
+
+你可以为一组较长的文本，添加多组向量，从而在检索时，只要其中一组向量被检索到，该数据也将被召回。
+
+## 提高向量搜索精度的方法
+
+1. 更好分词分段：当一段话的结构和语义是完整的，并且是单一的，精度也会提高。因此，许多系统都会优化分词器，尽可能的保障每组数据的完整性。
+2. 精简`index`的内容，减少向量内容的长度：当`index`的内容更少，更准确时，检索精度自然会提高。但与此同时，会牺牲一定的检索范围，适合答案较为严格的场景。
+3. 丰富`index`的数量，可以为同一个`chunk`内容增加多组`index`。
 4. 优化检索词：在实际使用过程中，用户的问题通常是模糊的或是缺失的，并不一定是完整清晰的问题。因此优化用户的问题（检索词）很大程度上也可以提高精度。
 5. 微调向量模型：由于市面上直接使用的向量模型都是通用型模型，在特定领域的检索精度并不高，因此微调向量模型可以很大程度上提高专业领域的检索效果。
 

packages/global/common/string/textSplitter.ts

@@ -63,8 +63,8 @@ export const splitText2Chunks = (props: { text: string; maxLen: number; overlapL
     let chunks: string[] = [];
     for (let i = 0; i < splitTexts.length; i++) {
       let text = splitTexts[i];
-      let chunkToken = countPromptTokens(lastChunk, '');
-      const textToken = countPromptTokens(text, '');
+      let chunkToken = lastChunk.length;
+      const textToken = text.length;
 
       // next chunk is too large / new chunk is too large(The current chunk must be smaller than maxLen)
       if (textToken >= maxLen || chunkToken + textToken > maxLen * 1.4) {

packages/global/core/dataset/type.d.ts

-import type { VectorModelItemType } from '../../core/ai/model.d';
+import type { LLMModelItemType, VectorModelItemType } from '../../core/ai/model.d';
 import { PermissionTypeEnum } from '../../support/permission/constant';
 import { PushDatasetDataChunkProps } from './api';
 import {
@@ -19,6 +19,7 @@ export type DatasetSchemaType = {
   avatar: string;
   name: string;
   vectorModel: string;
+  agentModel: string;
   tags: string[];
   type: `${DatasetTypeEnum}`;
   permission: `${PermissionTypeEnum}`;
@@ -84,8 +85,9 @@ export type CollectionWithDatasetType = Omit<DatasetCollectionSchemaType, 'datas
 };
 
 /* ================= dataset ===================== */
-export type DatasetItemType = Omit<DatasetSchemaType, 'vectorModel'> & {
+export type DatasetItemType = Omit<DatasetSchemaType, 'vectorModel' | 'agentModel'> & {
   vectorModel: VectorModelItemType;
+  agentModel: LLMModelItemType;
   isOwner: boolean;
   canWrite: boolean;
 };

packages/global/support/wallet/bill/api.d.ts

@@ -3,6 +3,8 @@ import { BillListItemType } from './type';
 
 export type CreateTrainingBillProps = {
   name: string;
+  vectorModel?: string;
+  agentModel?: string;
 };
 
 export type ConcatBillProps = {

packages/service/core/app/schema.ts

@@ -61,7 +61,6 @@ const AppSchema = new Schema({
 
 try {
   AppSchema.index({ updateTime: -1 });
-  AppSchema.index({ 'share.collection': -1 });
 } catch (error) {
   console.log(error);
 }

packages/service/core/dataset/collection/schema.ts

@@ -69,7 +69,6 @@ const DatasetCollectionSchema = new Schema({
 
 try {
   DatasetCollectionSchema.index({ datasetId: 1 });
-  DatasetCollectionSchema.index({ userId: 1 });
   DatasetCollectionSchema.index({ updateTime: -1 });
 } catch (error) {
   console.log(error);

packages/service/core/dataset/schema.ts

@@ -48,6 +48,11 @@ const DatasetSchema = new Schema({
     required: true,
     default: 'text-embedding-ada-002'
   },
+  agentModel: {
+    type: String,
+    required: true,
+    default: 'gpt-3.5-turbo-16k'
+  },
   type: {
     type: String,
     enum: Object.keys(DatasetTypeMap),

packages/service/core/dataset/training/schema.ts

@@ -95,7 +95,7 @@ const TrainingDataSchema = new Schema({
 
 try {
   TrainingDataSchema.index({ lockTime: 1 });
-  TrainingDataSchema.index({ userId: 1 });
+  TrainingDataSchema.index({ datasetId: 1 });
   TrainingDataSchema.index({ collectionId: 1 });
   TrainingDataSchema.index({ expireAt: 1 }, { expireAfterSeconds: 7 * 24 * 60 });
 } catch (error) {

projects/app/public/locales/en/common.json

@@ -250,6 +250,7 @@
     }
   },
   "dataset": {
+    "Agent Model": "Learning Model",
     "Chunk Length": "Chunk Length",
     "Confirm move the folder": "Confirm Move",
     "Confirm to delete the data": "Confirm to delete the data?",
@@ -259,6 +260,7 @@
     "Delete Dataset Error": "Delete dataset failed",
     "Edit Folder": "Edit Folder",
     "Export": "Export",
+    "Export Dataset Limit Error": "Export Data Error",
     "File Input": "Import File",
     "File Size": "File Size",
     "Filename": "Filename",

projects/app/public/locales/zh/common.json

@@ -250,6 +250,7 @@
     }
   },
   "dataset": {
+    "Agent Model": "文件处理模型",
     "Chunk Length": "数据总量",
     "Confirm move the folder": "确认移动到该目录",
     "Confirm to delete the data": "确认删除该数据？",
@@ -259,6 +260,7 @@
     "Delete Dataset Error": "删除知识库异常",
     "Edit Folder": "编辑文件夹",
     "Export": "导出",
+    "Export Dataset Limit Error": "导出数据失败",
     "File Input": "文件导入",
     "File Size": "文件大小",
     "Filename": "文件名",

projects/app/src/constants/dataset.ts

+import { defaultQAModels, defaultVectorModels } from '@fastgpt/global/core/ai/model';
 import type {
   DatasetCollectionItemType,
   DatasetItemType
@@ -17,13 +18,8 @@ export const defaultDatasetDetail: DatasetItemType = {
   permission: 'private',
   isOwner: false,
   canWrite: false,
-  vectorModel: {
-    model: 'text-embedding-ada-002',
-    name: 'Embedding-2',
-    price: 0.2,
-    defaultToken: 500,
-    maxToken: 3000
-  }
+  vectorModel: defaultVectorModels[0],
+  agentModel: defaultQAModels[0]
 };
 
 export const defaultCollectionDetail: DatasetCollectionItemType = {
@@ -43,7 +39,8 @@ export const defaultCollectionDetail: DatasetCollectionItemType = {
     name: '',
     tags: [],
     permission: 'private',
-    vectorModel: 'text-embedding-ada-002'
+    vectorModel: defaultVectorModels[0].model,
+    agentModel: defaultQAModels[0].model
   },
   parentId: '',
   name: '',

projects/app/src/global/core/api/datasetReq.d.ts

@@ -5,6 +5,7 @@ import type { SearchTestItemType } from '@/types/core/dataset';
 import { UploadChunkItemType } from '@fastgpt/global/core/dataset/type';
 import { DatasetCollectionSchemaType } from '@fastgpt/global/core/dataset/type';
 import { PermissionTypeEnum } from '@fastgpt/global/support/permission/constant';
+import type { LLMModelItemType } from '@fastgpt/global/core/ai/model.d';
 
 /* ===== dataset ===== */
 export type DatasetUpdateParams = {
@@ -14,6 +15,7 @@ export type DatasetUpdateParams = {
   name?: string;
   avatar?: string;
   permission?: `${PermissionTypeEnum}`;
+  agentModel?: LLMModelItemType;
 };
 
 export type SearchTestProps = {

projects/app/src/global/core/dataset/api.d.ts

@@ -9,6 +9,7 @@ export type CreateDatasetParams = {
   tags: string;
   avatar: string;
   vectorModel?: string;
+  agentModel?: string;
   type: `${DatasetTypeEnum}`;
 };
 

projects/app/src/global/core/prompt/agent.ts

 export const Prompt_AgentQA = {
   prompt: `我会给你一段文本，{{theme}}，学习它们，并整理学习成果，要求为：
-1. 提出最多 25 个问题。
-2. 给出每个问题的答案。
-3. 答案要详细完整，答案可以包含普通文字、链接、代码、表格、公示、媒体链接等 markdown 元素。
+1. 提出问题并给出每个问题的答案。
+2. 每个答案都要详细完整，给出相关原文描述，答案可以包含普通文字、链接、代码、表格、公示、媒体链接等 markdown 元素。
+3. 最多提出 30 个问题。
 4. 按格式返回多个问题和答案:
 
 Q1: 问题。

projects/app/src/pages/api/admin/initv46-2.ts

@@ -11,6 +11,8 @@ import {
 import { authCert } from '@fastgpt/service/support/permission/auth/common';
 import { MongoDatasetData } from '@fastgpt/service/core/dataset/data/schema';
 import { getUserDefaultTeam } from '@fastgpt/service/support/user/team/controller';
+import { MongoDataset } from '@fastgpt/service/core/dataset/schema';
+import { defaultQAModels } from '@fastgpt/global/core/ai/model';
 
 let success = 0;
 /* pg 中的数据搬到 mongo dataset.datas 中，并做映射 */
@@ -41,6 +43,13 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse)
 
     await initPgData();
 
+    await MongoDataset.updateMany(
+      {},
+      {
+        agentModel: defaultQAModels[0].model
+      }
+    );
+
     jsonRes(res, {
       data: await init(limit),
       message:
@@ -76,14 +85,19 @@ async function initPgData() {
   for (let i = 0; i < limit; i++) {
     init(i);
   }
+
   async function init(index: number): Promise<any> {
     const userId = rows[index]?.user_id;
     if (!userId) return;
     try {
       const tmb = await getUserDefaultTeam({ userId });
+      console.log(tmb);
+
       // update pg
       await PgClient.query(
-        `Update ${PgDatasetTableName} set team_id = '${tmb.teamId}', tmb_id = '${tmb.tmbId}' where user_id = '${userId}' AND team_id='null';`
+        `Update ${PgDatasetTableName} set team_id = '${String(tmb.teamId)}', tmb_id = '${String(
+          tmb.tmbId
+        )}' where user_id = '${userId}' AND team_id='null';`
       );
       console.log(++success);
       init(index + limit);

projects/app/src/pages/api/admin/initv46-3.ts

+import type { NextApiRequest, NextApiResponse } from 'next';
+import { jsonRes } from '@fastgpt/service/common/response';
+import { connectToDatabase } from '@/service/mongo';
+import { delay } from '@/utils/tools';
+import { PgClient } from '@fastgpt/service/common/pg';
+import {
+  DatasetDataIndexTypeEnum,
+  PgDatasetTableName
+} from '@fastgpt/global/core/dataset/constant';
+
+import { authCert } from '@fastgpt/service/support/permission/auth/common';
+import { MongoDatasetData } from '@fastgpt/service/core/dataset/data/schema';
+
+let success = 0;
+/* pg 中的数据搬到 mongo dataset.datas 中，并做映射 */
+export default async function handler(req: NextApiRequest, res: NextApiResponse) {
+  try {
+    const { limit = 50 } = req.body as { limit: number };
+    await authCert({ req, authRoot: true });
+    await connectToDatabase();
+    success = 0;
+
+    jsonRes(res, {
+      data: await init(limit)
+    });
+  } catch (error) {
+    console.log(error);
+
+    jsonRes(res, {
+      code: 500,
+      error
+    });
+  }
+}
+
+type PgItemType = {
+  id: string;
+  q: string;
+  a: string;
+  dataset_id: string;
+  collection_id: string;
+  data_id: string;
+};
+
+async function init(limit: number): Promise<any> {
+  const { rows: idList } = await PgClient.query<{ id: string }>(
+    `SELECT id FROM ${PgDatasetTableName} WHERE inited=1`
+  );
+
+  console.log('totalCount', idList.length);
+
+  await delay(2000);
+
+  if (idList.length === 0) return;
+
+  for (let i = 0; i < limit; i++) {
+    initData(i);
+  }
+
+  async function initData(index: number): Promise<any> {
+    const dataId = idList[index]?.id;
+    if (!dataId) {
+      console.log('done');
+      return;
+    }
+    // get limit data where data_id is null
+    const { rows } = await PgClient.query<PgItemType>(
+      `SELECT id,q,a,dataset_id,collection_id,data_id FROM ${PgDatasetTableName} WHERE id=${dataId};`
+    );
+    const data = rows[0];
+    if (!data) {
+      console.log('done');
+      return;
+    }
+
+    try {
+      // update mongo data and update inited
+      await MongoDatasetData.findByIdAndUpdate(data.data_id, {
+        q: data.q,
+        a: data.a,
+        indexes: [
+          {
+            defaultIndex: !data.a,
+            type: data.a ? DatasetDataIndexTypeEnum.qa : DatasetDataIndexTypeEnum.chunk,
+            dataId: data.id,
+            text: data.q
+          }
+        ]
+      });
+      // update pg data_id
+      await PgClient.query(`UPDATE ${PgDatasetTableName} SET inited=0 WHERE id=${dataId};`);
+
+      return initData(index + limit);
+    } catch (error) {
+      console.log(error);
+      console.log(data);
+      await delay(500);
+      return initData(index);
+    }
+  }
+}

projects/app/src/pages/api/core/dataset/allDataset.ts

@@ -2,7 +2,7 @@ import type { NextApiRequest, NextApiResponse } from 'next';
 import { jsonRes } from '@fastgpt/service/common/response';
 import { connectToDatabase } from '@/service/mongo';
 import { MongoDataset } from '@fastgpt/service/core/dataset/schema';
-import { getVectorModel } from '@/service/core/ai/model';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
 import type { DatasetItemType } from '@fastgpt/global/core/dataset/type.d';
 import { mongoRPermission } from '@fastgpt/global/support/permission/utils';
 import { authUserRole } from '@fastgpt/service/support/permission/auth/user';
@@ -22,6 +22,7 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
     const data = datasets.map((item) => ({
       ...item.toJSON(),
       vectorModel: getVectorModel(item.vectorModel),
+      agentModel: getQAModel(item.agentModel),
       canWrite: String(item.tmbId) === tmbId,
       isOwner: teamOwner || String(item.tmbId) === tmbId
     }));

projects/app/src/pages/api/core/dataset/checkExportLimit.ts

+import type { NextApiRequest, NextApiResponse } from 'next';
+import { jsonRes } from '@fastgpt/service/common/response';
+import { connectToDatabase } from '@/service/mongo';
+import { MongoUser } from '@fastgpt/service/support/user/schema';
+import { addLog } from '@fastgpt/service/common/mongo/controller';
+import { authDataset } from '@fastgpt/service/support/permission/auth/dataset';
+import { MongoDatasetData } from '@fastgpt/service/core/dataset/data/schema';
+import { findDatasetIdTreeByTopDatasetId } from '@fastgpt/service/core/dataset/controller';
+
+export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
+  try {
+    await connectToDatabase();
+    let { datasetId } = req.query as {
+      datasetId: string;
+    };
+
+    if (!datasetId) {
+      throw new Error('缺少参数');
+    }
+
+    // 凭证校验
+    const { userId } = await authDataset({ req, authToken: true, datasetId, per: 'w' });
+
+    await limitCheck({
+      datasetId,
+      userId
+    });
+
+    jsonRes(res);
+  } catch (err) {
+    res.status(500);
+    jsonRes(res, {
+      code: 500,
+      error: err
+    });
+  }
+}
+
+export async function limitCheck({ datasetId, userId }: { datasetId: string; userId: string }) {
+  const exportIds = await findDatasetIdTreeByTopDatasetId(datasetId);
+
+  const limitMinutesAgo = new Date(
+    Date.now() - (global.feConfigs?.limit?.exportLimitMinutes || 0) * 60 * 1000
+  );
+
+  // auth export times
+  const authTimes = await MongoUser.findOne(
+    {
+      _id: userId,
+      $or: [
+        { 'limit.exportKbTime': { $exists: false } },
+        { 'limit.exportKbTime': { $lte: limitMinutesAgo } }
+      ]
+    },
+    '_id limit'
+  );
+
+  if (!authTimes) {
+    const minutes = `${global.feConfigs?.limit?.exportLimitMinutes || 0} 分钟`;
+    return Promise.reject(`上次导出未到 ${minutes}，每 ${minutes}仅可导出一次。`);
+  }
+
+  // auth max data
+  const total = await MongoDatasetData.countDocuments({
+    datasetId: { $in: exportIds }
+  });
+
+  addLog.info(`export datasets: ${datasetId}`, { total });
+
+  if (total > 100000) {
+    return Promise.reject('数据量超出 10 万，无法导出');
+  }
+}

projects/app/src/pages/api/core/dataset/create.ts

@@ -9,7 +9,8 @@ import { authUserNotVisitor } from '@fastgpt/service/support/permission/auth/use
 export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
   try {
     await connectToDatabase();
-    const { name, tags, avatar, vectorModel, parentId, type } = req.body as CreateDatasetParams;
+    const { name, tags, avatar, vectorModel, agentModel, parentId, type } =
+      req.body as CreateDatasetParams;
 
     // 凭证校验
     const { teamId, tmbId } = await authUserNotVisitor({ req, authToken: true });
@@ -20,6 +21,7 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
       tmbId,
       tags,
       vectorModel,
+      agentModel,
       avatar,
       parentId: parentId || null,
       type

projects/app/src/pages/api/core/dataset/data/pushData.ts

@@ -10,7 +10,7 @@ import { countPromptTokens } from '@fastgpt/global/common/string/tiktoken';
 import type { PushDataResponse } from '@/global/core/api/datasetRes.d';
 import type { PushDatasetDataProps } from '@/global/core/dataset/api.d';
 import { PushDatasetDataChunkProps } from '@fastgpt/global/core/dataset/api';
-import { getVectorModel } from '@/service/core/ai/model';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
 import { authDatasetCollection } from '@fastgpt/service/support/permission/auth/dataset';
 import { getCollectionWithDataset } from '@fastgpt/service/core/dataset/controller';
 
@@ -63,24 +63,14 @@ export async function pushDataToDatasetCollection({
   mode,
   prompt,
   billId
-}: { teamId: string; tmbId: string } & PushDatasetDataProps): Promise<PushDataResponse> {
-  // get dataset vector model
-  const {
-    datasetId: { _id: datasetId, vectorModel }
-  } = await getCollectionWithDataset(collectionId);
-
-  const vectorModelData = getVectorModel(vectorModel);
-
-  const modeMap = {
-    [TrainingModeEnum.chunk]: {
-      maxToken: vectorModelData.maxToken * 1.5,
-      model: vectorModelData.model
-    },
-    [TrainingModeEnum.qa]: {
-      maxToken: global.qaModels[0].maxContext * 0.8,
-      model: global.qaModels[0].model
-    }
-  };
+}: {
+  teamId: string;
+  tmbId: string;
+} & PushDatasetDataProps): Promise<PushDataResponse> {
+  const { datasetId, model, maxToken } = await checkModelValid({
+    mode,
+    collectionId
+  });
 
   // filter repeat or equal content
   const set = new Set();
@@ -102,12 +92,13 @@ export async function pushDataToDatasetCollection({
       // count q token
       const token = countPromptTokens(item.q);
 
-      if (token > modeMap[mode].maxToken) {
+      if (token > maxToken) {
         filterResult.overToken.push(item);
         return;
       }
 
       if (set.has(text)) {
+        console.log('repeat', item);
         filterResult.repeat.push(item);
       } else {
         filterResult.success.push(item);
@@ -126,7 +117,7 @@ export async function pushDataToDatasetCollection({
       billId,
       mode,
       prompt,
-      model: modeMap[mode].model,
+      model,
       q: item.q,
       a: item.a,
       indexes: item.indexes
@@ -142,6 +133,44 @@ export async function pushDataToDatasetCollection({
   };
 }
 
+export async function checkModelValid({
+  mode,
+  collectionId
+}: {
+  mode: `${TrainingModeEnum}`;
+  collectionId: string;
+}) {
+  const {
+    datasetId: { _id: datasetId, vectorModel, agentModel }
+  } = await getCollectionWithDataset(collectionId);
+
+  if (mode === TrainingModeEnum.chunk) {
+    if (!collectionId) return Promise.reject(`CollectionId is empty`);
+    const vectorModelData = getVectorModel(vectorModel);
+    if (!vectorModelData) {
+      return Promise.reject(`Model ${vectorModel} is inValid`);
+    }
+    return {
+      datasetId,
+      maxToken: vectorModelData.maxToken * 1.5,
+      model: vectorModelData.model
+    };
+  }
+
+  if (mode === TrainingModeEnum.qa) {
+    const qaModelData = getQAModel(agentModel);
+    if (!qaModelData) {
+      return Promise.reject(`Model ${agentModel} is inValid`);
+    }
+    return {
+      datasetId,
+      maxToken: qaModelData.maxContext * 0.8,
+      model: qaModelData.model
+    };
+  }
+  return Promise.reject(`Mode ${mode} is inValid`);
+}
+
 export const config = {
   api: {
     bodyParser: {

projects/app/src/pages/api/core/dataset/detail.ts

 import type { NextApiRequest, NextApiResponse } from 'next';
 import { jsonRes } from '@fastgpt/service/common/response';
 import { connectToDatabase } from '@/service/mongo';
-import { getVectorModel } from '@/service/core/ai/model';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
 import type { DatasetItemType } from '@fastgpt/global/core/dataset/type.d';
 import { authDataset } from '@fastgpt/service/support/permission/auth/dataset';
 
@@ -28,6 +28,7 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
       data: {
         ...dataset,
         vectorModel: getVectorModel(dataset.vectorModel),
+        agentModel: getQAModel(dataset.agentModel),
         canWrite,
         isOwner
       }

projects/app/src/pages/api/core/dataset/exportAll.ts

 import type { NextApiRequest, NextApiResponse } from 'next';
-import { jsonRes } from '@fastgpt/service/common/response';
+import { jsonRes, responseWriteController } from '@fastgpt/service/common/response';
 import { connectToDatabase } from '@/service/mongo';
 import { MongoUser } from '@fastgpt/service/support/user/schema';
 import { addLog } from '@fastgpt/service/common/mongo/controller';
@@ -8,6 +8,7 @@ import { MongoDatasetData } from '@fastgpt/service/core/dataset/data/schema';
 import { findDatasetIdTreeByTopDatasetId } from '@fastgpt/service/core/dataset/controller';
 import { Readable } from 'stream';
 import type { Cursor } from '@fastgpt/service/common/mongo';
+import { limitCheck } from './checkExportLimit';
 
 export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
   try {
@@ -23,39 +24,12 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
     // 凭证校验
     const { userId } = await authDataset({ req, authToken: true, datasetId, per: 'w' });
 
-    const exportIds = await findDatasetIdTreeByTopDatasetId(datasetId);
-
-    const limitMinutesAgo = new Date(
-      Date.now() - (global.feConfigs?.limit?.exportLimitMinutes || 0) * 60 * 1000
-    );
-
-    // auth export times
-    const authTimes = await MongoUser.findOne(
-      {
-        _id: userId,
-        $or: [
-          { 'limit.exportKbTime': { $exists: false } },
-          { 'limit.exportKbTime': { $lte: limitMinutesAgo } }
-        ]
-      },
-      '_id limit'
-    );
-
-    if (!authTimes) {
-      const minutes = `${global.feConfigs?.limit?.exportLimitMinutes || 0} 分钟`;
-      throw new Error(`上次导出未到 ${minutes}，每 ${minutes}仅可导出一次。`);
-    }
-
-    // auth max data
-    const total = await MongoDatasetData.countDocuments({
-      datasetId: { $in: exportIds }
+    await limitCheck({
+      userId,
+      datasetId
     });
 
-    addLog.info(`export datasets: ${datasetId}`, { total });
-
-    if (total > 100000) {
-      throw new Error('数据量超出 10 万，无法导出');
-    }
+    const exportIds = await findDatasetIdTreeByTopDatasetId(datasetId);
 
     res.setHeader('Content-Type', 'text/csv; charset=utf-8;');
     res.setHeader('Content-Disposition', 'attachment; filename=dataset.csv; ');
@@ -72,35 +46,27 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
       'q a'
     ).cursor();
 
-    function cursorToReadableStream(cursor: Cursor) {
-      const readable = new Readable({
-        objectMode: true,
-        read() {}
-      });
+    const write = responseWriteController({
+      res,
+      readStream: cursor
+    });
 
-      readable.push(`\uFEFFindex,content`);
+    write(`\uFEFFindex,content`);
 
-      cursor.on('data', (doc) => {
-        const q = doc.q.replace(/"/g, '""') || '';
-        const a = doc.a.replace(/"/g, '""') || '';
+    cursor.on('data', (doc) => {
+      const q = doc.q.replace(/"/g, '""') || '';
+      const a = doc.a.replace(/"/g, '""') || '';
 
-        readable.push(`\n"${q}","${a}"`);
-      });
+      write(`\n"${q}","${a}"`);
+    });
 
-      cursor.on('end', async () => {
-        readable.push(null);
-        cursor.close();
-        await MongoUser.findByIdAndUpdate(userId, {
-          'limit.exportKbTime': new Date()
-        });
+    cursor.on('end', async () => {
+      cursor.close();
+      res.end();
+      await MongoUser.findByIdAndUpdate(userId, {
+        'limit.exportKbTime': new Date()
       });
-
-      return readable;
-    }
-
-    // @ts-ignore
-    const stream = cursorToReadableStream(cursor);
-    stream.pipe(res);
+    });
   } catch (err) {
     res.status(500);
     jsonRes(res, {

projects/app/src/pages/api/core/dataset/list.ts

 import type { NextApiRequest, NextApiResponse } from 'next';
 import { jsonRes } from '@fastgpt/service/common/response';
 import { connectToDatabase } from '@/service/mongo';
-import { getVectorModel } from '@/service/core/ai/model';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
 import type { DatasetItemType } from '@fastgpt/global/core/dataset/type.d';
 import { DatasetTypeEnum } from '@fastgpt/global/core/dataset/constant';
 import { MongoDataset } from '@fastgpt/service/core/dataset/schema';
@@ -28,6 +28,7 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
       datasets.map(async (item) => ({
         ...item.toJSON(),
         vectorModel: getVectorModel(item.vectorModel),
+        agentModel: getQAModel(item.agentModel),
         canWrite,
         isOwner: teamOwner || String(item.tmbId) === tmbId
       }))

projects/app/src/pages/api/core/dataset/update.ts

@@ -8,7 +8,8 @@ import { authDataset } from '@fastgpt/service/support/permission/auth/dataset';
 export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
   try {
     await connectToDatabase();
-    const { id, parentId, name, avatar, tags, permission } = req.body as DatasetUpdateParams;
+    const { id, parentId, name, avatar, tags, permission, agentModel } =
+      req.body as DatasetUpdateParams;
 
     if (!id) {
       throw new Error('缺少参数');
@@ -26,7 +27,8 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
         ...(name && { name }),
         ...(avatar && { avatar }),
         ...(tags && { tags }),
-        ...(permission && { permission })
+        ...(permission && { permission }),
+        ...(agentModel && { agentModel: agentModel.model })
       }
     );
 

projects/app/src/pages/api/support/wallet/bill/createTrainingBill.ts

@@ -5,15 +5,17 @@ import { MongoBill } from '@fastgpt/service/support/wallet/bill/schema';
 import { authCert } from '@fastgpt/service/support/permission/auth/common';
 import { BillSourceEnum } from '@fastgpt/global/support/wallet/bill/constants';
 import { CreateTrainingBillProps } from '@fastgpt/global/support/wallet/bill/api.d';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
 
 export default async function handler(req: NextApiRequest, res: NextApiResponse) {
   try {
     await connectToDatabase();
-    const { name } = req.body as CreateTrainingBillProps;
+    const { name, vectorModel, agentModel } = req.body as CreateTrainingBillProps;
 
     const { teamId, tmbId } = await authCert({ req, authToken: true, authApiKey: true });
 
-    const qaModel = global.qaModels[0];
+    const vectorModelData = getVectorModel(vectorModel);
+    const agentModelData = getQAModel(agentModel);
 
     const { _id } = await MongoBill.create({
       teamId,
@@ -23,13 +25,13 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse)
       list: [
         {
           moduleName: '索引生成',
-          model: 'embedding',
+          model: vectorModelData.name,
           amount: 0,
           tokenLen: 0
         },
         {
           moduleName: 'QA 拆分',
-          model: qaModel?.name,
+          model: agentModelData.name,
           amount: 0,
           tokenLen: 0
         }

projects/app/src/pages/dataset/detail/components/DataCard.tsx

@@ -170,7 +170,7 @@ const DataCard = () => {
       </Flex>
       <Grid
         minH={'100px'}
-        gridTemplateColumns={['1fr', 'repeat(2,1fr)', 'repeat(3,1fr)']}
+        gridTemplateColumns={['1fr', 'repeat(2,1fr)', 'repeat(3,1fr)', 'repeat(4,1fr)']}
         gridGap={4}
       >
         {datasetDataList.map((item) => (

projects/app/src/pages/dataset/detail/components/Import/ImportModal.tsx

@@ -34,10 +34,10 @@ const ImportData = ({
   const theme = useTheme();
   const { datasetDetail } = useDatasetStore();
   const [importType, setImportType] = useState<`${ImportTypeEnum}`>(ImportTypeEnum.chunk);
+  const vectorModel = datasetDetail.vectorModel;
+  const agentModel = datasetDetail.agentModel;
 
   const typeMap = useMemo(() => {
-    const vectorModel = datasetDetail.vectorModel;
-    const qaModel = qaModelList[0];
     const map = {
       [ImportTypeEnum.chunk]: {
         defaultChunkLen: vectorModel?.defaultToken || 500,
@@ -45,8 +45,8 @@ const ImportData = ({
         mode: TrainingModeEnum.chunk
       },
       [ImportTypeEnum.qa]: {
-        defaultChunkLen: qaModel?.maxContext * 0.5 || 8000,
-        unitPrice: qaModel?.price || 3,
+        defaultChunkLen: agentModel?.maxContext * 0.6 || 9000,
+        unitPrice: agentModel?.price || 3,
         mode: TrainingModeEnum.qa
       },
       [ImportTypeEnum.csv]: {
@@ -56,7 +56,13 @@ const ImportData = ({
       }
     };
     return map[importType];
-  }, [datasetDetail.vectorModel, importType]);
+  }, [
+    agentModel?.maxContext,
+    agentModel?.price,
+    importType,
+    vectorModel?.defaultToken,
+    vectorModel?.price
+  ]);
 
   const TitleStyle: BoxProps = {
     fontWeight: 'bold',
@@ -104,8 +110,10 @@ const ImportData = ({
 
         <Provider
           {...typeMap}
+          vectorModel={vectorModel.model}
+          agentModel={agentModel.model}
+          datasetId={datasetDetail._id}
           importType={importType}
-          datasetId={datasetId}
           parentId={parentId}
           onUploadSuccess={uploadSuccess}
         >

projects/app/src/pages/dataset/detail/components/Import/Provider.tsx

@@ -90,6 +90,8 @@ const Provider = ({
   parentId,
   unitPrice,
   mode,
+  vectorModel,
+  agentModel,
   defaultChunkLen = 500,
   importType,
   onUploadSuccess,
@@ -99,6 +101,8 @@ const Provider = ({
   parentId: string;
   unitPrice: number;
   mode: `${TrainingModeEnum}`;
+  vectorModel: string;
+  agentModel: string;
   defaultChunkLen: number;
   importType: `${ImportTypeEnum}`;
   onUploadSuccess: () => void;
@@ -132,7 +136,9 @@ const Provider = ({
         const chunks = file.chunks;
         // create training bill
         const billId = await postCreateTrainingBill({
-          name: t('dataset.collections.Create Training Data', { filename: file.filename })
+          name: t('dataset.collections.Create Training Data', { filename: file.filename }),
+          vectorModel,
+          agentModel
         });
         // create a file collection and training bill
         const collectionId = await postDatasetCollection({

projects/app/src/pages/dataset/detail/components/Import/QA.tsx

@@ -13,8 +13,8 @@ const fileExtension = '.txt, .doc, .docx, .pdf, .md';
 
 const QAImport = () => {
   const { datasetDetail } = useDatasetStore();
-  const vectorModel = datasetDetail.vectorModel;
-  const unitPrice = vectorModel?.price || 0.2;
+  const agentModel = datasetDetail.agentModel;
+  const unitPrice = agentModel?.price || 3;
 
   const {
     successChunks,

projects/app/src/pages/dataset/detail/components/Info.tsx

@@ -9,7 +9,7 @@ import React, {
 import { useRouter } from 'next/router';
 import { Box, Flex, Button, FormControl, IconButton, Input } from '@chakra-ui/react';
 import { QuestionOutlineIcon, DeleteIcon } from '@chakra-ui/icons';
-import { delDatasetById, putDatasetById } from '@/web/core/dataset/api';
+import { delDatasetById } from '@/web/core/dataset/api';
 import { useSelectFile } from '@/web/common/file/hooks/useSelectFile';
 import { useToast } from '@/web/common/hooks/useToast';
 import { useDatasetStore } from '@/web/core/dataset/store/dataset';
@@ -22,6 +22,8 @@ import Tag from '@/components/Tag';
 import MyTooltip from '@/components/MyTooltip';
 import { useTranslation } from 'next-i18next';
 import PermissionRadio from '@/components/support/permission/Radio';
+import MySelect from '@/components/Select';
+import { qaModelList } from '@/web/common/system/staticData';
 
 export interface ComponentRef {
   initInput: (tags: string) => void;
@@ -50,7 +52,7 @@ const Info = (
     multiple: false
   });
 
-  const { datasetDetail, loadDatasetDetail, loadDatasets } = useDatasetStore();
+  const { datasetDetail, loadDatasetDetail, loadDatasets, updateDataset } = useDatasetStore();
 
   /* 点击删除 */
   const onclickDelKb = useCallback(async () => {
@@ -76,11 +78,10 @@ const Info = (
     async (data: DatasetItemType) => {
       setBtnLoading(true);
       try {
-        await putDatasetById({
+        await updateDataset({
           id: datasetId,
           ...data
         });
-        await loadDatasetDetail(datasetId, true);
         toast({
           title: '更新成功',
           status: 'success'
@@ -94,7 +95,7 @@ const Info = (
       }
       setBtnLoading(false);
     },
-    [loadDatasetDetail, datasetId, loadDatasets, toast]
+    [updateDataset, datasetId, loadDatasetDetail, toast, loadDatasets]
   );
   const saveSubmitError = useCallback(() => {
     // deep search message
@@ -194,6 +195,27 @@ const Info = (
           })}
         />
       </FormControl>
+      <Flex mt={6} alignItems={'center'}>
+        <Box flex={['0 0 90px', '0 0 160px']} w={0}>
+          {t('dataset.Agent Model')}
+        </Box>
+        <Box flex={[1, '0 0 300px']}>
+          <MySelect
+            w={'100%'}
+            value={getValues('agentModel').model}
+            list={qaModelList.map((item) => ({
+              label: item.name,
+              value: item.model
+            }))}
+            onchange={(e) => {
+              const agentModel = qaModelList.find((item) => item.model === e);
+              if (!agentModel) return;
+              setValue('agentModel', agentModel);
+              setRefresh((state) => !state);
+            }}
+          />
+        </Box>
+      </Flex>
       <Flex mt={8} alignItems={'center'} w={'100%'} flexWrap={'wrap'}>
         <Box flex={['0 0 90px', '0 0 160px']} w={0}>
           标签

projects/app/src/pages/dataset/detail/components/InputDataModal.tsx

@@ -196,7 +196,7 @@ const InputDataModal = ({
   const loading = useMemo(() => isImporting || isUpdating, [isImporting, isUpdating]);
 
   return (
-    <MyModal isOpen={true} isCentered w={'90vw'} maxW={'90vw'} h={'90vh'}>
+    <MyModal isOpen={true} isCentered w={'90vw'} maxW={'1440px'} h={'90vh'}>
       <Flex h={'100%'}>
         <Box p={5} borderRight={theme.borders.base}>
           <RawSourceText
@@ -250,7 +250,7 @@ const InputDataModal = ({
                     mt={1}
                     placeholder={`该输入框是必填项，该内容通常是对于知识点的描述，也可以是用户的问题，最多 ${maxToken} 字。`}
                     maxLength={maxToken}
-                    rows={10}
+                    rows={12}
                     bg={'myWhite.400'}
                     {...register(`q`, {
                       required: true
@@ -274,7 +274,7 @@ const InputDataModal = ({
                       maxToken * 1.5
                     } 字。`}
                     bg={'myWhite.400'}
-                    rows={10}
+                    rows={12}
                     maxLength={maxToken * 1.5}
                     {...register('a')}
                   />

projects/app/src/pages/dataset/list/component/CreateModal.tsx

@@ -15,10 +15,12 @@ import { postCreateDataset } from '@/web/core/dataset/api';
 import type { CreateDatasetParams } from '@/global/core/dataset/api.d';
 import MySelect from '@/components/Select';
 import { QuestionOutlineIcon } from '@chakra-ui/icons';
-import { vectorModelList } from '@/web/common/system/staticData';
+import { vectorModelList, qaModelList } from '@/web/common/system/staticData';
 import Tag from '@/components/Tag';
+import { useTranslation } from 'next-i18next';
 
 const CreateModal = ({ onClose, parentId }: { onClose: () => void; parentId?: string }) => {
+  const { t } = useTranslation();
   const [refresh, setRefresh] = useState(false);
   const { toast } = useToast();
   const router = useRouter();
@@ -29,6 +31,7 @@ const CreateModal = ({ onClose, parentId }: { onClose: () => void; parentId?: st
       name: '',
       tags: '',
       vectorModel: vectorModelList[0].model,
+      agentModel: qaModelList[0].model,
       type: 'dataset',
       parentId
     }
@@ -76,7 +79,7 @@ const CreateModal = ({ onClose, parentId }: { onClose: () => void; parentId?: st
   });
 
   return (
-    <MyModal isOpen onClose={onClose} isCentered={!isPc} w={'400px'}>
+    <MyModal isOpen onClose={onClose} isCentered={!isPc} w={'450px'}>
       <ModalHeader fontSize={'2xl'}>创建一个知识库</ModalHeader>
       <ModalBody>
         <Box color={'myGray.800'} fontWeight={'bold'}>
@@ -106,7 +109,7 @@ const CreateModal = ({ onClose, parentId }: { onClose: () => void; parentId?: st
           />
         </Flex>
         <Flex mt={6} alignItems={'center'}>
-          <Box flex={'0 0 80px'}>索引模型</Box>
+          <Box flex={'0 0 100px'}>索引模型</Box>
           <Box flex={1}>
             <MySelect
               w={'100%'}
@@ -122,8 +125,25 @@ const CreateModal = ({ onClose, parentId }: { onClose: () => void; parentId?: st
             />
           </Box>
         </Flex>
+        <Flex mt={6} alignItems={'center'}>
+          <Box flex={'0 0 100px'}>{t('dataset.Agent Model')}</Box>
+          <Box flex={1}>
+            <MySelect
+              w={'100%'}
+              value={getValues('agentModel')}
+              list={qaModelList.map((item) => ({
+                label: item.name,
+                value: item.model
+              }))}
+              onchange={(e) => {
+                setValue('agentModel', e);
+                setRefresh((state) => !state);
+              }}
+            />
+          </Box>
+        </Flex>
         <Flex mt={6} alignItems={'center'} w={'100%'}>
-          <Box flex={'0 0 80px'}>
+          <Box flex={'0 0 100px'}>
             标签
             <MyTooltip label={'用空格隔开多个标签，便于搜索'} forceShow>
               <QuestionOutlineIcon ml={1} />

projects/app/src/pages/dataset/list/index.tsx

@@ -20,7 +20,8 @@ import {
   delDatasetById,
   getDatasetPaths,
   putDatasetById,
-  postCreateDataset
+  postCreateDataset,
+  getCheckExportLimit
 } from '@/web/core/dataset/api';
 import { useTranslation } from 'next-i18next';
 import Avatar from '@/components/Avatar';
@@ -38,6 +39,7 @@ import { useDrag } from '@/web/common/hooks/useDrag';
 import { useUserStore } from '@/web/support/user/useUserStore';
 import PermissionIconText from '@/components/support/permission/IconText';
 import { PermissionTypeEnum } from '@fastgpt/global/support/permission/constant';
+import { DatasetItemType } from '@fastgpt/global/core/dataset/type';
 
 const CreateModal = dynamic(() => import('./component/CreateModal'), { ssr: false });
 const MoveModal = dynamic(() => import('./component/MoveModal'), { ssr: false });
@@ -89,6 +91,23 @@ const Kb = () => {
     successToast: t('common.Delete Success'),
     errorToast: t('dataset.Delete Dataset Error')
   });
+  // check export limit
+  const { mutate: exportDataset } = useRequest({
+    mutationFn: async (dataset: DatasetItemType) => {
+      setLoading(true);
+      await getCheckExportLimit(dataset._id);
+      const a = document.createElement('a');
+      a.href = `/api/core/dataset/exportAll?datasetId=${dataset._id}`;
+      a.download = `${dataset.name}.csv`;
+      document.body.appendChild(a);
+      a.click();
+      document.body.removeChild(a);
+    },
+    onSettled() {
+      setLoading(false);
+    },
+    errorToast: t('dataset.Export Dataset Limit Error')
+  });
 
   const { data, refetch } = useQuery(['loadDataset', parentId], () => {
     return Promise.all([loadDatasets(parentId), getDatasetPaths(parentId)]);
@@ -371,12 +390,7 @@ const Kb = () => {
                       </Flex>
                     ),
                     onClick: () => {
-                      const a = document.createElement('a');
-                      a.href = `/api/core/dataset/exportAll?datasetId=${dataset._id}`;
-                      a.download = `${dataset.name}.csv`;
-                      document.body.appendChild(a);
-                      a.click();
-                      document.body.removeChild(a);
+                      exportDataset(dataset);
                     }
                   },
                   {

projects/app/src/service/events/generateQA.ts

@@ -109,6 +109,7 @@ export async function generateQA(): Promise<any> {
 
   try {
     const startTime = Date.now();
+    const model = data.model ?? global.qaModels[0].model;
 
     // request LLM to get QA
     const messages: ChatMessageItemType[] = [
@@ -122,9 +123,10 @@ export async function generateQA(): Promise<any> {
             })
       }
     ];
-    const ai = getAIApi(undefined, 480000);
+
+    const ai = getAIApi(undefined, 600000);
     const chatResponse = await ai.chat.completions.create({
-      model: global.qaModels[0].model,
+      model,
       temperature: 0.01,
       messages,
       stream: false
@@ -147,8 +149,11 @@ export async function generateQA(): Promise<any> {
     // delete data from training
     await MongoDatasetTraining.findByIdAndDelete(data._id);
 
-    console.log(`split result length: `, qaArr.length);
-    console.log('生成QA成功，time:', `${(Date.now() - startTime) / 1000}s`);
+    addLog.info(`QA Training Finish`, {
+      time: `${(Date.now() - startTime) / 1000}s`,
+      splitLength: qaArr.length,
+      usage: chatResponse.usage
+    });
 
     // add bill
     if (qaArr.length > 0) {
@@ -156,7 +161,8 @@ export async function generateQA(): Promise<any> {
         teamId: data.teamId,
         tmbId: data.tmbId,
         totalTokens,
-        billId: data.billId
+        billId: data.billId,
+        model
       });
     } else {
       addLog.info(`QA result 0:`, { answer });

projects/app/src/service/support/wallet/bill/push.ts

 import { BillSourceEnum } from '@fastgpt/global/support/wallet/bill/constants';
-import { getAudioSpeechModel } from '@/service/core/ai/model';
+import { getAudioSpeechModel, getQAModel } from '@/service/core/ai/model';
 import type { ChatHistoryItemResType } from '@fastgpt/global/core/chat/api.d';
 import { formatPrice } from '@fastgpt/global/support/wallet/bill/tools';
 import { addLog } from '@fastgpt/service/common/mongo/controller';
@@ -9,10 +9,16 @@ import { POST } from '@fastgpt/service/common/api/plusRequest';
 
 export function createBill(data: CreateBillProps) {
   if (!global.systemEnv.pluginBaseUrl) return;
+  if (data.total === 0) {
+    addLog.info('0 Bill', data);
+  }
   POST('/support/wallet/bill/createBill', data);
 }
 export function concatBill(data: ConcatBillProps) {
   if (!global.systemEnv.pluginBaseUrl) return;
+  if (data.total === 0) {
+    addLog.info('0 Bill', data);
+  }
   POST('/support/wallet/bill/concatBill', data);
 }
 
@@ -59,18 +65,18 @@ export const pushChatBill = ({
 export const pushQABill = async ({
   teamId,
   tmbId,
+  model,
   totalTokens,
   billId
 }: {
   teamId: string;
   tmbId: string;
+  model: string;
   totalTokens: number;
   billId: string;
 }) => {
-  addLog.info('splitData generate success', { totalTokens });
-
   // 获取模型单价格
-  const unitPrice = global.qaModels?.[0]?.price || 3;
+  const unitPrice = getQAModel(model).price;
   // 计算价格
   const total = unitPrice * totalTokens;
 

projects/app/src/web/core/dataset/api.ts

@@ -48,6 +48,9 @@ export const putDatasetById = (data: DatasetUpdateParams) => PUT(`/core/dataset/
 
 export const delDatasetById = (id: string) => DELETE(`/core/dataset/delete?id=${id}`);
 
+export const getCheckExportLimit = (datasetId: string) =>
+  GET(`/core/dataset/checkExportLimit`, { datasetId });
+
 /* =========== search test ============ */
 export const postSearchText = (data: SearchTestProps) =>
   POST<SearchDataResponseItemType[]>(`/core/dataset/searchTest`, data);