1. 概述

基于LangChain實現(xiàn)的零代碼平臺，可以通過拖拽的方式輕松實現(xiàn)LLM Apps.
FlowiseAI/Flowise: Drag & drop UI to build your customized LLM flow
Getting Started - FlowiseAI

2. LLM Apps

2.1. 基于本地LLM的翻譯應(yīng)用

2.2. 基于本地LLM的聊天機器人

2.3. RAG應(yīng)用

存在一個問題尚未解決：
https://github.com/lm-sys/FastChat/issues/2336

3. 如何部署

3.1. FastChat

lm-sys/FastChat: An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena

3.2 Milvus

Install Milvus Standalone with Docker Compose Milvus documentation

• docker-compose.yml

version: '3.5'

services:
  etcd:
    container_name: milvus-etcd
    image: quay.io/coreos/etcd:v3.5.5
    environment:
      - ETCD_AUTO_COMPACTION_MODE=revision
      - ETCD_AUTO_COMPACTION_RETENTION=1000
      - ETCD_QUOTA_BACKEND_BYTES=4294967296
      - ETCD_SNAPSHOT_COUNT=50000
    volumes:
      - ${DOCKER_VOLUME_DIRECTORY:-.}/volumes/etcd:/etcd
    command: etcd -advertise-client-urls=http://127.0.0.1:2379 -listen-client-urls http://0.0.0.0:2379 --data-dir /etcd
    healthcheck:
      test: ["CMD", "etcdctl", "endpoint", "health"]
      interval: 30s
      timeout: 20s
      retries: 3

  minio:
    container_name: milvus-minio
    image: minio/minio:RELEASE.2023-03-20T20-16-18Z
    environment:
      MINIO_ACCESS_KEY: minioadmin
      MINIO_SECRET_KEY: minioadmin
    ports:
      - "9001:9001"
      - "9000:9000"
    volumes:
      - ${DOCKER_VOLUME_DIRECTORY:-.}/volumes/minio:/minio_data
    command: minio server /minio_data --console-address ":9001"
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:9000/minio/health/live"]
      interval: 30s
      timeout: 20s
      retries: 3

  standalone:
    container_name: milvus-standalone
    image: milvusdb/milvus:v2.3.10
    command: ["milvus", "run", "standalone"]
    security_opt:
    - seccomp:unconfined
    environment:
      ETCD_ENDPOINTS: etcd:2379
      MINIO_ADDRESS: minio:9000
    volumes:
      - ${DOCKER_VOLUME_DIRECTORY:-.}/volumes/milvus:/var/lib/milvus
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:9091/healthz"]
      interval: 30s
      start_period: 90s
      timeout: 20s
      retries: 3
    ports:
      - "19530:19530"
      - "9091:9091"
    depends_on:
      - "etcd"
      - "minio"

networks:
  default:
    name: milvus

3.3. Flowise

• docker-compose.yaml

version: '3.4'

services:
  mysql:
    container_name: mysql
    restart: always
    image: mysql:8.0
    environment:
      - "MYSQL_ROOT_PASSWORD=1234"  # 數(shù)據(jù)庫密碼，建議修改，如果修改需要同步修改bisheng/congfig/config.yaml配置
      - "MYSQL_DATABASE=flowise"
      - "TZ=Asia/Shanghai"
    ports:
      - "3306:3306"
    volumes:
      - ${DOCKER_VOLUME_DIRECTORY:-.}/mysql/my.cnf:/etc/mysql/my.cnf
      - ${DOCKER_VOLUME_DIRECTORY:-.}/data/mysql:/var/lib/mysql

  flowise:
    container_name: flowise
    image: flowiseai/flowise
    restart: always
    environment:
        - PORT=3000
        - FLOWISE_USERNAME=admin
        - FLOWISE_PASSWORD=1234
        - DEBUG=true
        - DATABASE_PATH=/data
        - DATABASE_TYPE=mysql
        - DATABASE_PORT=3306
        - DATABASE_HOST=mysql
        - DATABASE_NAME=flowise
        - DATABASE_USER=root
        - DATABASE_PASSWORD=1234
        - APIKEY_PATH=/data
        - SECRETKEY_PATH=/data
        - FLOWISE_SECRETKEY_OVERWRITE=myencryptionkey
        - LOG_LEVEL=debug
        - LOG_PATH=/data/log
    ports:
        - '3000:3000'
    volumes:
        - ./data/flowise:/data
    command: /bin/sh -c "sleep 3; flowise start"
    depends_on:
      - "mysql"

• my.cnf

[client]
default-character-set=utf8mb4

[mysql]
default-character-set=utf8mb4

[mysqld]
# 解決 node 的兼容問題
default_authentication_plugin=mysql_native_password
init_connect='SET collation_connection = utf8mb4_unicode_ci, NAMES utf8mb4'
character-set-server=utf8mb4
collation-server=utf8mb4_unicode_ci
# skip-character-set-client-handshake
sql_mode=STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_ENGINE_SUBSTITUTION

4. 后續(xù)

Flowise和LangChain是兩個基于LLM（大型語言模型）的應(yīng)用構(gòu)建工具。本文主要介紹了如何利用Flowise和LangChain快速構(gòu)建LLM應(yīng)用。

通過結(jié)合Flowise和LangChain，開發(fā)者可以快速構(gòu)建基于LLM的自然語言處理應(yīng)用，實現(xiàn)文本分析、問答、文本生成等功能。

• Flowise提供了易用的界面和豐富的API接口，幫助開發(fā)者快速將自然語言處理任務(wù)集成到應(yīng)用中。
• LangChain提供了多語言支持和統(tǒng)一的平臺，幫助開發(fā)者在多個語言和領(lǐng)域中實現(xiàn)自然語言處理應(yīng)用。

結(jié)合Flowise和LangChain，開發(fā)者可以更加高效地構(gòu)建基于LLM的自然語言處理應(yīng)用，為用戶提供更加智能和便捷的服務(wù)。

Flowise｜無代碼 ChatBot 構(gòu)建平臺｜LangChain - 知乎 (zhihu.com)
flowise-docker-compose: flowise-docker-compose (gitee.com)