v0.4.0

✨ Enhancements

• Refactor DB-GPT Web
• Add DB-GPT MetaData Module (#696)
• Switch BaseDao from MetaData (#693)
• Add ChatPlugin Module (#680)
• Support use custom log file name and tracer file name (#690)
• Read and save system information to tracer (#670)
• Support cross-service data recording and analyze (#665)
• Support vLLM (#658)
• Llama.cpp support new GGUF file format (#649)

🐞 Bug fixes

• Fix remote embedding model start error (#708)
• DB metadata charset bug (#698)
• ChatData (#694)
• Fix type exception when use office ppt database (#679)
• DB schema not complete bug (#675)
• Fix load some quantization model error (#674)
• Fix openai adapt previous proxy_server_url configuration and support azure openai model (#668)
• use openai package instead of requests (#619)
• Fix zh_CN document error for Debugging (#666)
• Chroma persist_path compatible with old version (#657)
• .env other vector db customize vector store config does not work (#656)

🛠️ Other improvements

• Update tutorial and application document
• Add unit test for worker manager (#681)
• Fix zh_CN document error for Debugging (#666)

Thank you to all our contributors for making this release possible!
@Aries-ckt, @SimonChuZZ, @cabljac, @csunny, @fangyinc, @oushu1zhangxiangxuan1, @peter-wangxu, @quqibing, @yhjun1026, @yihong0618 and simonchuzz

v0.3.9

✨ Enhancements

• Modify PR release drafter (#639)
• Add Postgresql support (#626)
• Add custom text separators and refactor log configuration (#636)
• Update Spark datasource (#631)
• Add model web management (#613)
• Add incremental response to streaming response for /v1/chat/completion request (#611)
• Multi-module dependency splitting (#606)
• New worker manager factory componet (#600)
• Prompt manage support (#599)

🐞 Bug fixes

• Fix MySQL connection bug (#640)
• Use thread pool to replace native threading.Thread (#628)
• Fix starting the service due to network proxy.(#616)
• ChatExcel OutParse Bug Fix (#596)

🛠️ Other improvements

• Roll back json parsing code (#637)

Thank you to all our contributors for making this release possible!
@Aries-ckt, @cm-liushaodong, @csunny, @fangyinc, @lozzo, @yhjun1026, @yihong0618 and lozzow

v0.3.8

✨ Enhancements

• Support openai embedding model (#591)
• Support InternLM (#583)
• Multi LLM switch (#576)
• Multi-model supports embedding model and simple component design implementation (#574)
• Supports the deployment of multiple models through the API and add the corresponding command line interface (#570)
• Unified Deployment Mode with multi-model and add command line chat with model (#566)
• Support Baichuan2 model (#558)
• Milvus support collection delete and delete by ids (#550)
• Multi-model support with proxyllm and add more command-cli (#532)

🐞 Bug fixes

• ChatExcel bug fix (#593)
• ChatExcel OutParse Bug Fix (#588)
• Fix remote embedding model error in some case (#587)
• Fix(dashboard):get report parameter (#586)
• Fix(web):web proxy address (#582)
• Fix proxy llm with fix model from CONFIG (#568)

🛠️ Other improvements

• Latest AutoDL image (#595)
• Add pr labeler and release drafter workflows (#559)
• New issue template and release drafter (#549)
• Command-line tool with knowledge repository initialization (#521)

Thank you to all our contributors for making this release possible!
@Aries-ckt, @csunny, @fangyinc, @junewgl and @yhjun1026

v0.3.7

What's changed

1. Add ChatExcel @yhjun1026 #495
2. Add DBEditor @yhjun1026 #495
3. Add LLM Model Management @fangyinc #501
4. Multi Datasource Management @yhjun1026 #495
5. Add bge Embedding Model @Aries-ckt #479
6. Fix gpt4all output bug @csunny #478

新特性

1. 新增ChatExcel @yhjun1026 #495
2. 新增DB Sql编辑器 @yhjun1026 #495
3. 新增多模型管理 @fangyinc #501
4. 新增多数据源管理 @yhjun1026 #495
5. 新增bge Embedding 模型 @Aries-ckt #479
6. 解决gpt4all 输出问题 @csunny #478

v0.3.6

What's changed
1.Sqlite as default DB @fangyinc #438
2.LLM Model support llama.cpp @fangyinc #452
3.Docker Image CICD @fangyinc #466
4.Refactor DB-GPT Document @Aries-ckt #465
5.Fix upload file 405 error @Aries-ckt #435

新特性
1.Sqlite作为默认数据库类型 @fangyinc #438
2.LLM Model 支持 llama.cpp @fangyinc #452
3.Docker镜像CICD支持 @fangyinc #466
4.DB-GPT 文档重构 @Aries-ckt #465
5.修复上传文件 405 error @Aries-ckt #435

v0.3.5

What's changed
1.KBQA Support Arguments @Aries-ckt #426
2.Docker actions build @yihong0618 #422

新特性
1.KBQA 支持调参 @Aries-ckt #426
2.Docker 构建自动化 @yihong0618 #422

v0.3.4

What's changed

1.Add Multi-LLMs support, vicuna-v1.5 and WizardLM-v1.2 @fangyinc
2.Support 8-bit quantization and 4-bit quantization @fangyinc
3.Multi DB connection @yhjun1026
4.Support Bard Proxy @xuyuan23
5.knowledge space and document delete @Aries-ckt

新特性

1.支持vicuna-v1.5 和 WizardLM-v1.2 模型 @fangyinc
2.支持8-bit和4-bit 量化版本 @fangyinc
3.多数据库连接支持 @yhjun1026 @HPc369
4.Bard代理支持 @xuyuan23
5. 知识库空间和文档删除 @Aries-ckt @ssw1999

v0.3.3

What's changed

• Add Multi-LLMs support, LLaMA-2(7b, 13b, 70b), BaiChuan(7b, 13b) @fangyinc
• Fix DML and DDL Execute bug. @yhjun1026
• Chat2Word support @Aries-ckt
• Split Webui to DB-GPT-Web subprojet @yhjun1026
• Fix MySql Keyword bug @yiqijiu
• Add Multi-Database connections. (MySQL, mssql, etc) @yhjun1026
• Add Readme Content. @csunny

v0.3.2

What's Changed

1. support llama-2 model @fangyinc
2. add autoDL image and faq @Aries-ckt
3. add dbgpt api and you can use dbgpt according pip install dbgpt @Aries-ckt
4. structured prompt data interaction between dbgpt-server and llm-server @fangyinc

新特性

1. 支持llama-2系列模型
2. 提供AutoDL镜像
3. dbgpt api v0.3.0
4. 结构化prompt

v0.3.1

What's Changed

1. Solving the chatglm2-6b Multi GPUS Reasoning Problem @qutcat1997
2. chatglm not working in doc qa, meta qa and plugin @fangyinc

新特性

1. 解决chatglm2-6b多卡推理问题 @qutcat1997
2. chatglm 模型在知识问答, 数据库问答优化 @fangyinc