feat(model): multi-model supports embedding model and simple componen…

…t design implementation

docker/compose_examples/cluster-docker-compose.yml

@@ -4,9 +4,10 @@ services:
   controller:
     image: eosphorosai/dbgpt:latest
     command: dbgpt start controller
+    restart: unless-stopped
     networks:
       - dbgptnet
-  worker:
+  llm-worker:
     image: eosphorosai/dbgpt:latest
     command: dbgpt start worker --model_name vicuna-13b-v1.5 --model_path /app/models/vicuna-13b-v1.5 --port 8001 --controller_addr http://controller:8000
     environment:
@@ -17,6 +18,27 @@ services:
       - /data:/data
       # Please modify it to your own model directory
       - /data/models:/app/models
+    restart: unless-stopped
+    networks:
+      - dbgptnet
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              capabilities: [gpu]
+  embedding-worker:
+    image: eosphorosai/dbgpt:latest
+    command: dbgpt start worker --model_name text2vec --worker_type text2vec --model_path /app/models/text2vec-large-chinese --port 8002 --controller_addr http://controller:8000
+    environment:
+      - DBGPT_LOG_LEVEL=DEBUG
+    depends_on:
+      - controller
+    volumes:
+      - /data:/data
+      # Please modify it to your own model directory
+      - /data/models:/app/models
+    restart: unless-stopped
     networks:
       - dbgptnet
     deploy:
@@ -37,7 +59,8 @@ services:
       - MODEL_SERVER=http://controller:8000
     depends_on:
       - controller
-      - worker
+      - llm-worker
+      - embedding-worker
     volumes:
       - /data:/data
       # Please modify it to your own model directory

docs/getting_started/install.rst

@@ -9,6 +9,7 @@ DB-GPT product is a Web application that you can chat database, chat knowledge,
 - docker
 - docker_compose
 - environment
+- cluster deployment
 - deploy_faq
 
 .. toctree::
@@ -20,6 +21,7 @@ DB-GPT product is a Web application that you can chat database, chat knowledge,
    ./install/deploy/deploy.md
    ./install/docker/docker.md
    ./install/docker_compose/docker_compose.md
+   ./install/cluster/cluster.rst
    ./install/llm/llm.rst
    ./install/environment/environment.md
    ./install/faq/deploy_faq.md

docs/getting_started/install/cluster/cluster.rst

@@ -0,0 +1,19 @@
+Cluster deployment
+==================================
+
+In order to deploy DB-GPT to multiple nodes, you can deploy a cluster. The cluster architecture diagram is as follows:
+
+.. raw:: html
+
+    <img src="../../../_static/img/muti-model-cluster-overview.png" />
+
+
+* On :ref:`Deploying on local machine <local-cluster-index>`. Local cluster deployment.
+
+.. toctree::
+   :maxdepth: 2
+   :caption: Cluster deployment
+   :name: cluster_deploy
+   :hidden:
+
+   ./vms/index.md

docs/getting_started/install/cluster/kubernetes/index.md

@@ -0,0 +1,3 @@
+Kubernetes cluster deployment
+==================================
+(kubernetes-cluster-index)=

docs/getting_started/install/llm/cluster/model_cluster.md → docs/getting_started/install/cluster/vms/index.md

@@ -1,6 +1,6 @@
-Cluster deployment
+Local cluster deployment
 ==================================
-
+(local-cluster-index)=
 ## Model cluster deployment
 
 
@@ -17,7 +17,7 @@ dbgpt start controller
 By default, the Model Controller starts on port 8000.
 
 
-### Launch Model Worker
+### Launch LLM Model Worker
 
 If you are starting `chatglm2-6b`:
 
@@ -39,6 +39,18 @@ dbgpt start worker --model_name vicuna-13b-v1.5 \
 
 Note: Be sure to use your own model name and model path.
 
+### Launch Embedding Model Worker
+
+```bash
+
+dbgpt start worker --model_name text2vec \
+--model_path /app/models/text2vec-large-chinese \
+--worker_type text2vec \
+--port 8003 \
+--controller_addr http://127.0.0.1:8000
+```
+
+Note: Be sure to use your own model name and model path.
 
 Check your model:
 
@@ -51,8 +63,12 @@ You will see the following output:
 +-----------------+------------+------------+------+---------+---------+-----------------+----------------------------+
 |    Model Name   | Model Type |    Host    | Port | Healthy | Enabled | Prompt Template |       Last Heartbeat       |
 +-----------------+------------+------------+------+---------+---------+-----------------+----------------------------+
-|   chatglm2-6b   |    llm     | 172.17.0.6 | 8001 |   True  |   True  |       None      | 2023-08-31T04:48:45.252939 |
-| vicuna-13b-v1.5 |    llm     | 172.17.0.6 | 8002 |   True  |   True  |       None      | 2023-08-31T04:48:55.136676 |
+|   chatglm2-6b   |    llm     | 172.17.0.2 | 8001 |   True  |   True  |                 | 2023-09-12T23:04:31.287654 |
+|  WorkerManager  |  service   | 172.17.0.2 | 8001 |   True  |   True  |                 | 2023-09-12T23:04:31.286668 |
+|  WorkerManager  |  service   | 172.17.0.2 | 8003 |   True  |   True  |                 | 2023-09-12T23:04:29.845617 |
+|  WorkerManager  |  service   | 172.17.0.2 | 8002 |   True  |   True  |                 | 2023-09-12T23:04:24.598439 |
+|     text2vec    |  text2vec  | 172.17.0.2 | 8003 |   True  |   True  |                 | 2023-09-12T23:04:29.844796 |
+| vicuna-13b-v1.5 |    llm     | 172.17.0.2 | 8002 |   True  |   True  |                 | 2023-09-12T23:04:24.597775 |
 +-----------------+------------+------------+------+---------+---------+-----------------+----------------------------+
 ```
 
@@ -69,15 +85,15 @@ MODEL_SERVER=http://127.0.0.1:8000
 #### Start the webserver
 
 ```bash
-python pilot/server/dbgpt_server.py --light
+dbgpt start webserver --light
 ```
 
 `--light`  indicates not to start the embedded model service.
 
 Alternatively, you can prepend the command with `LLM_MODEL=chatglm2-6b` to start:
 
 ```bash
-LLM_MODEL=chatglm2-6b python pilot/server/dbgpt_server.py --light
+LLM_MODEL=chatglm2-6b dbgpt start webserver --light
 ```
 
 
@@ -101,9 +117,11 @@ Options:
   --help            Show this message and exit.
 
 Commands:
-  model  Clients that manage model serving
-  start  Start specific server.
-  stop   Start specific server.
+  install    Install dependencies, plugins, etc.
+  knowledge  Knowledge command line tool
+  model      Clients that manage model serving
+  start      Start specific server.
+  stop       Start specific server.
 ```
 
 **View the `dbgpt start` help**
@@ -146,10 +164,11 @@ Options:
   --model_name TEXT               Model name  [required]
   --model_path TEXT               Model path  [required]
   --worker_type TEXT              Worker type
-  --worker_class TEXT             Model worker class, pilot.model.worker.defau
-                                  lt_worker.DefaultModelWorker
+  --worker_class TEXT             Model worker class,
+                                  pilot.model.cluster.DefaultModelWorker
   --host TEXT                     Model worker deploy host  [default: 0.0.0.0]
-  --port INTEGER                  Model worker deploy port  [default: 8000]
+  --port INTEGER                  Model worker deploy port  [default: 8001]
+  --daemon                        Run Model Worker in background
   --limit_model_concurrency INTEGER
                                   Model concurrency limit  [default: 5]
   --standalone                    Standalone mode. If True, embedded Run
@@ -166,7 +185,7 @@ Options:
                                   (seconds)  [default: 20]
   --device TEXT                   Device to run model. If None, the device is
                                   automatically determined
-  --model_type TEXT               Model type, huggingface or llama.cpp
+  --model_type TEXT               Model type, huggingface, llama.cpp and proxy
                                   [default: huggingface]
   --prompt_template TEXT          Prompt template. If None, the prompt
                                   template is automatically determined from
@@ -190,7 +209,7 @@ Options:
   --compute_dtype TEXT            Model compute type
   --trust_remote_code             Trust remote code  [default: True]
   --verbose                       Show verbose output.
-  --help                          Show this message and exit.
+  --help                          Show this message and exit. 
 ```
 
 **View the `dbgpt model`help**
@@ -208,10 +227,13 @@ Usage: dbgpt model [OPTIONS] COMMAND [ARGS]...
 
 Options:
   --address TEXT  Address of the Model Controller to connect to. Just support
-                  light deploy model  [default: http://127.0.0.1:8000]
+                  light deploy model, If the environment variable
+                  CONTROLLER_ADDRESS is configured, read from the environment
+                  variable
   --help          Show this message and exit.
 
 Commands:
+  chat     Interact with your bot from the command line
   list     List model instances
   restart  Restart model instances
   start    Start model instances

docs/getting_started/install/llm/llm.rst

@@ -6,6 +6,7 @@ DB-GPT provides a management and deployment solution for multiple models. This c
 
 
 Multi LLMs Support, Supports multiple large language models, currently supporting
+  - 🔥 Baichuan2(7b,13b)
   - 🔥 Vicuna-v1.5(7b,13b)
   - 🔥 llama-2(7b,13b,70b)
   - WizardLM-v1.2(13b)
@@ -19,7 +20,6 @@ Multi LLMs Support, Supports multiple large language models, currently supportin
 
 - llama_cpp
 - quantization
-- cluster deployment
 
 .. toctree::
    :maxdepth: 2
@@ -29,4 +29,3 @@ Multi LLMs Support, Supports multiple large language models, currently supportin
 
    ./llama/llama_cpp.md
    ./quantization/quantization.md
-   ./cluster/model_cluster.md

docs/locales/zh_CN/LC_MESSAGES/getting_started/install.po

@@ -8,7 +8,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: DB-GPT 👏👏 0.3.5\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2023-08-16 18:31+0800\n"
+"POT-Creation-Date: 2023-09-13 09:06+0800\n"
 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language: zh_CN\n"
@@ -19,34 +19,38 @@ msgstr ""
 "Content-Transfer-Encoding: 8bit\n"
 "Generated-By: Babel 2.12.1\n"
 
-#: ../../getting_started/install.rst:2 ../../getting_started/install.rst:14
-#: 2861085e63144eaca1bb825e5f05d089
+#: ../../getting_started/install.rst:2 ../../getting_started/install.rst:15
+#: e2c13385046b4da6b6838db6ba2ea59c
 msgid "Install"
 msgstr "Install"
 
-#: ../../getting_started/install.rst:3 01a6603d91fa4520b0f839379d4eda23
+#: ../../getting_started/install.rst:3 3cb6cd251ed440dabe5d4f556435f405
 msgid ""
 "DB-GPT product is a Web application that you can chat database, chat "
 "knowledge, text2dashboard."
 msgstr "DB-GPT 可以生成sql，智能报表, 知识库问答的产品"
 
-#: ../../getting_started/install.rst:8 beca85cddc9b4406aecf83d5dfcce1f7
+#: ../../getting_started/install.rst:8 6fe8104b70d24f5fbfe2ad9ebf3bc3ba
 msgid "deploy"
 msgstr "部署"
 
-#: ../../getting_started/install.rst:9 601e9b9eb91f445fb07d2f1c807f0370
+#: ../../getting_started/install.rst:9 e67974b3672346809febf99a3b9a55d3
 msgid "docker"
 msgstr "docker"
 
-#: ../../getting_started/install.rst:10 6d1e094ac9284458a32a3e7fa6241c81
+#: ../../getting_started/install.rst:10 64de16a047c74598966e19a656bf6c4f
 msgid "docker_compose"
 msgstr "docker_compose"
 
-#: ../../getting_started/install.rst:11 ff1d1c60bbdc4e8ca82b7a9f303dd167
+#: ../../getting_started/install.rst:11 9f87d65e8675435b87cb9376a5bfd85c
 msgid "environment"
 msgstr "environment"
 
-#: ../../getting_started/install.rst:12 33bfbe8defd74244bfc24e8fbfd640f6
+#: ../../getting_started/install.rst:12 e60fa13bb24544ed9d4f902337093ebc
+msgid "cluster deployment"
+msgstr "集群部署"
+
+#: ../../getting_started/install.rst:13 7451712679c2412e858e7d3e2af6b174
 msgid "deploy_faq"
 msgstr "deploy_faq"
 

docs/locales/zh_CN/LC_MESSAGES/getting_started/install/cluster/cluster.po

@@ -0,0 +1,42 @@
+# SOME DESCRIPTIVE TITLE.
+# Copyright (C) 2023, csunny
+# This file is distributed under the same license as the DB-GPT package.
+# FIRST AUTHOR <EMAIL@ADDRESS>, 2023.
+#
+#, fuzzy
+msgid ""
+msgstr ""
+"Project-Id-Version: DB-GPT 👏👏 0.3.6\n"
+"Report-Msgid-Bugs-To: \n"
+"POT-Creation-Date: 2023-09-13 10:11+0800\n"
+"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
+"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
+"Language: zh_CN\n"
+"Language-Team: zh_CN <[email protected]>\n"
+"Plural-Forms: nplurals=1; plural=0;\n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=utf-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+"Generated-By: Babel 2.12.1\n"
+
+#: ../../getting_started/install/cluster/cluster.rst:2
+#: ../../getting_started/install/cluster/cluster.rst:13
+#: 69804208b580447798d6946150da7bdf
+msgid "Cluster deployment"
+msgstr "集群部署"
+
+#: ../../getting_started/install/cluster/cluster.rst:4
+#: fa3e4e0ae60a45eb836bcd256baa9d91
+msgid ""
+"In order to deploy DB-GPT to multiple nodes, you can deploy a cluster. "
+"The cluster architecture diagram is as follows:"
+msgstr "为了能将 DB-GPT 部署到多个节点上，你可以部署一个集群，集群的架构图如下:"
+
+#: ../../getting_started/install/cluster/cluster.rst:11
+#: e739449099ca43cabe9883233ca7e572
+#, fuzzy
+msgid ""
+"On :ref:`Deploying on local machine <local-cluster-index>`. Local cluster"
+" deployment."
+msgstr "关于 :ref:`在本地机器上部署 <local-cluster-index>`。本地集群部署。"
+

docs/locales/zh_CN/LC_MESSAGES/getting_started/install/cluster/kubernetes/index.po

@@ -0,0 +1,26 @@
+# SOME DESCRIPTIVE TITLE.
+# Copyright (C) 2023, csunny
+# This file is distributed under the same license as the DB-GPT package.
+# FIRST AUTHOR <EMAIL@ADDRESS>, 2023.
+#
+#, fuzzy
+msgid ""
+msgstr ""
+"Project-Id-Version: DB-GPT 👏👏 0.3.6\n"
+"Report-Msgid-Bugs-To: \n"
+"POT-Creation-Date: 2023-09-13 09:06+0800\n"
+"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
+"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
+"Language: zh_CN\n"
+"Language-Team: zh_CN <[email protected]>\n"
+"Plural-Forms: nplurals=1; plural=0;\n"
+"MIME-Version: 1.0\n"
+"Content-Type: text/plain; charset=utf-8\n"
+"Content-Transfer-Encoding: 8bit\n"
+"Generated-By: Babel 2.12.1\n"
+
+#: ../../getting_started/install/cluster/kubernetes/index.md:1
+#: 48e6f08f27c74f31a8b12758fe33dc24
+msgid "Kubernetes cluster deployment"
+msgstr "Kubernetes 集群部署"
+