v0.2.19

Highlights

• New topics in doc:
  • Prompting & reasoning: ReAct
  • Prompting & reasoning: Reflection
  • Task decomposition
• Improvement to code executors.
• Hotfix to address new numpy prerelease issue.
• Bug fix in AutoGenBench.

Thanks to @qingyun-wu @jackgerrits @davorrunje @lalo and all the other contributors!

What's Changed

• Require numpy be a 1.x release by @jackgerrits in #1966
• Fixes LFS problem with conversable-agent.jpg by @davorrunje in #1974
• Remove system message from code executors by @jackgerrits in #1933
• bump version to 0.2.19 and update doc by @sonichi in #1971
• Remove lint silencing comment by @jackgerrits in #1970
• Examples by topics by @qingyun-wu in #1967
• Add git-lhs to devcontainers by @davorrunje in #1979
• fix: [autogenbench] fails on win-os unless we specify encoding by @lalo in #1957
• Update local cli executor to use same filename strategy as docker by @jackgerrits in #1981

New Contributors

• @lalo made their first contribution in #1957

Full Changelog: v0.2.18...v0.2.19

v0.2.18

Highlights

• Support callable message: Example
• Adding a tool for fine tuning custom models on conversable agents
• Better command line code executor based on Docker
• Important doc updates:
  • Tutorial
  • Notebooks
• Bug fix for llm_config, powershell, logging etc.

Thanks to @qingyun-wu @olgavrou @jackgerrits @ekzhu @kevin666aa @rickyloynd-microsoft @cheng-tan @bassmang @WaelKarkoub @RohitRathore1 @bmuskalla @andreyseas @abhaymathur21 and all the other contributors!

What's Changed

• Update README of AutoGenBench MATH scenario by @kevin666aa in #1891
• Fix printout check for num code blocks by @jackgerrits in #1902
• Compressible Agent require model field for llm_config by @kevin666aa in #1903
• Implement docker based command line code executor by @jackgerrits in #1856
• Accept path for work_dir in LocalCommandLineCodeExecutor by @jackgerrits in #1909
• Add warning and reorder jupyter executor docs by @jackgerrits in #1908
• Fix broken links by @rickyloynd-microsoft in #1905
• Resolve all type issues in cache module by @jackgerrits in #1888
• Fix threading issue for logging by @cheng-tan in #1901
• Fix formatting in sequential chat notebook by @jackgerrits in #1914
• fix: minor fixes and formatting by @bassmang in #1919
• Log warning if DiskCache when RedisCache requested by @jackgerrits in #1921
• Add initial type check CI by @jackgerrits in #1882
• General Enhancements in agentchat 2.0 by @WaelKarkoub in #1906
• Adjusts send_introductions Type Hint in GroupChat by @WaelKarkoub in #1912
• AutoGen Tutorial by @ekzhu in #1702
• Fix remove -> remote by @jackgerrits in #1930
• Supporting callable message by @qingyun-wu in #1852
• Update more notebooks to be available on the website by @jackgerrits in #1890
• Update introduction.ipynb by @sonichi in #1934
• Update conversable agent image by @ekzhu in #1939
• Update tutorial examples, reference to register_reply. by @ekzhu in #1941
• Made a configurable constant by @RohitRathore1 in #1819
• Fix branch info for docker build by @bmuskalla in #1935
• Move LLM Caching docs to topics by @jackgerrits in #1950
• Update documentation for Quarto Version by @kevin666aa in #1951
• Fixed simple grammatical mistake by @andreyseas in #1952
• Update whisper notebook to use new function registration syntax by @jackgerrits in #1918
• In code_utils.py, handling NotADirectoryError as well in get_powershell_command by @abhaymathur21 in #1948
• Ability to fine tune custom model on conversable agents by @olgavrou in #1787
• improve validation of llm_config by @sonichi in #1946
• Make notebooks available in navbar by @jackgerrits in #1956

New Contributors

• @bassmang made their first contribution in #1919
• @bmuskalla made their first contribution in #1935
• @andreyseas made their first contribution in #1952

Full Changelog: v0.2.17...v0.2.18

v0.2.17

Highlights

• Summary of recent updates.
• Support customized speaker selection method: example.
• Improvement in nested chats and code execution.
• Improvement in doc, notebooks and docker file.
• Bug fix for clear history and custom client.
• Fix message processing order for proper combination of agent capabilities.

Thanks to @kevin666aa @ekzhu @jackgerrits @GregorD1A1 @KazooTTT @swiecki @truebit and all the other contributors!

What's Changed

• Remove default trigger value for register_nested_chats by @ekzhu in #1833
• Split apart ecosystem page, update sidebar, other website tweaks by @jackgerrits in #1812
• Upgrade Quarto and use notebook metadata for frontmatter by @jackgerrits in #1836
• Fix for bugs appeared when using function calls with clear history functionality by @GregorD1A1 in #1531
• revert quarto upgrade by @jackgerrits in #1840
• adjust the order of message processing by @sonichi in #1841
• add doc about effects for capabilities by @sonichi in #1842
• Upgrade quarto and include mitigation for title issue, fix links by @jackgerrits in #1851
• Create topics dir and move llm config by @jackgerrits in #1853
• Raise errors if incompatible args are used with an executor by @jackgerrits in #1855
• Add documentation for jupyter code executor by @jackgerrits in #1809
• summary of recent updates by @sonichi in #1850
• Small edits to update blog post by @jackgerrits in #1869
• Fix docs for. EmbeddedIPythonCodeExecutor by @jackgerrits in #1871
• Resolve warning on port binding by @jackgerrits in #1870
• Move executor diag print to conversable_agent by @jackgerrits in #1867
• Change name from LocalCommandlineCodeExecutor to LocalCommandLineCodeExecutor by @jackgerrits in #1873
• fix: update the docs link by @KazooTTT in #1879
• Fix devcontainer Dockerfile for arm by @jackgerrits in #1866
• Resolve type issues in redis cache by @jackgerrits in #1872
• UPDATE: Fixed broken examples links by @swiecki in #1886
• Add test to confirm that executor timeout preserves kernel state by @jackgerrits in #1885
• Add annoucement bar for what's new blog post by @jackgerrits in #1884
• fix ModelClient create function signature error by @truebit in #1865
• Fix issue with git checkout main by @ekzhu in #1876
• Download lfs in website build by @jackgerrits in #1894
• Allow user to pass in a customized speaker selection method by @kevin666aa in #1791

New Contributors

• @swiecki made their first contribution in #1886
• @truebit made their first contribution in #1865

Full Changelog: v0.2.16...v0.2.17

v0.2.16

Highlights

• Nested chats are made easier by register_nested_chats. Nested chats allow AutoGen agents to use other agents as their inner monologue before replying to other agents. Example of using nested chat to compose a super-agent based on multi-agent conversation.
• Blog for FSM-based group chat with user-specified agent transitions.
• Docker-based jupyter executor.
• Support removing function calls in ConversableAgent.
• Improvement in hook methods.
• Bug fix for DALLE agent.
• Improvement in doc & test.

Thanks to @qingyun-wu @joshkyh @freedeaths @jackgerrits @skzhang1 @RohitRathore1 @BeibinLi @shreyas36 @gunnarku @abhaymathur21 @victordibia and all the other contributors!

What's Changed

• Update agentchat_custom_model.ipynb by @shreyas90999 in #1768
• Blog regarding finite-state machine by @joshkyh in #1622
• Adding register_nested_chats to make nested chat easier by @qingyun-wu in #1753
• Support functions removing in ConversableAgent by @skzhang1 in #1786
• Don't require notebooks to have all imports at top by @jackgerrits in #1803
• hook-list are updated to accept a sender argument by @RohitRathore1 in #1799
• make default model of a constant class variable by @RohitRathore1 in #1780
• Debug: Resolve the DALLE Agent's self.client error. by @BeibinLi in #1798
• Add shebang functionality to tests by @gunnarku in #1784
• Implement docker based jupyter executor by @jackgerrits in #1794
• Add function generator and validator to gallery by @abhaymathur21 in #1807
• Extend process_notebooks for testing by @jackgerrits in #1789
• Update bug_report.yml [Invite users to explicitly state what model they are using that resulted in the issue] by @victordibia in #1811
• bug fix and cleanup for 0.2.16 by @sonichi in #1827

New Contributors

• @shreyas90999 made their first contribution in #1768

Full Changelog: v0.2.15...v0.2.16

v0.2.15

Highlights

• Async version of multiple dependent chats. Example.
• Improvement in chat control:
  • Allow sending introductions in the beginning of group chat for participants to know each other's role.
  • Allow setting max turns when initiating chats.
• Improvement and bug fix in:
  • custom message processing methods: allow processing messages before sending, such as displaying in a custom frontend.
  • multimodal agent: use PIL image internally.
  • code execution: command line executor, powershell etc.
  • long context handling.
  • GPT Assistant Agent: compatibility with azure openai.
  • AutoGenBench.
  • Documentation.

Thanks to @randombet @afourney @qingyun-wu @BeibinLi @jackgerrits @abhaymathur21 @skzhang1 @gunnarku @AaronWard @thinkall @dkirsche @RohitRathore1 @LinxinS97 @IANTHEREAL and all the other contributors!

What's Changed

• update ecosystem by @skzhang1 in #1624
• Check for missing dependencies before building the website by @gunnarku in #1678
• Allow limiting the maximum number of turns in initiate_chat and initiate_chats by @qingyun-wu in #1703
• Update example page by @qingyun-wu in #1698
• use str for hook key by @sonichi in #1711
• Add agent robot example to gallery by @AaronWard in #1718
• Use PIL Image internally for the Multimodal Agent by @BeibinLi in #1124
• Fix issue 1440 by applying new function registration decorator by @thinkall in #1661
• Command line code sanitation by @AaronWard in #1627
• news update by @sonichi in #1720
• fix web formats by @skzhang1 in #1728
• Updated code_utils.py & local_commandline_code_executor.py (powershell to pwsh) by @abhaymathur21 in #1710
• fix some docstring issues affecting rendering by @jackgerrits in #1739
• Refactor transform_messages by @dkirsche in #1631
• Async version of multiple sequential chat by @randombet in #1724
• Allow None for sender field in CoversableAgent.generate_reply by @RohitRathore1 in #1725
• [AutoBuild] fix test error by @LinxinS97 in #1750
• Updating code_utils.py to solve issue #1747 by @abhaymathur21 in #1758
• Add sidebar for notebooks page by @jackgerrits in #1766
• Use jupyer-kernel-gateway for ipython executor by @jackgerrits in #1748
• Handle azure_deployment Parameter Issue in GPTAssistantAgent to Maintain Compatibility with OpenAIWrapper by @IANTHEREAL in #1721
• Groupchat send introductions by @afourney in #961
• Version 0.0.2 of Autogenbench by @afourney in #1548
• process message before send by @sonichi in #1783

New Contributors

• @dkirsche made their first contribution in #1631
• @RohitRathore1 made their first contribution in #1725

Full Changelog: v0.2.14...v0.2.15

v0.2.14

Highlights

• Enhancement to sequential chats programming
  • support custom summary method
  • allow the chats to be initiated by different agents
  • example
• Improvement to GPTAssistantAgent
  • respect termination and human input mode
  • support Azure assistant API
• Runtime logging is back and advanced! Example
• Improvement to group chat: get nested agents and look up by name
• Doc improvement and bug fix.

Thanks to @qingyun-wu @yousonnet @IANTHEREAL @cheng-tan @WaelKarkoub @jackgerrits @bobbravo2 @maxim-saplin @olgavrou @gagb @FarshidShafia @gunnarku @Xtrah and all the other contributors!

What's Changed

• Rewrite and consolidate configuration docs by @jackgerrits in #1581
• Adding callable summary_method support and enhancements to initiate_chats by @qingyun-wu in #1628
• remove print config list by @sonichi in #1637
• return None instead of tuple in _generate_oai_reply_from_client by @sonichi in #1644
• [README] remove duplicated line by @bobbravo2 in #1646
• add autogen.initiate_chats by @qingyun-wu in #1638
• add GPTAssistantAgent is_termination_msg valid by @yousonnet in #1642
• FAQ, highlight the correct package name is pyautogen by @maxim-saplin in #1665
• Update gallery grid to flow better across screen sizes by @jackgerrits in #1652
• update dotnet workflow by @LittleLittleCloud in #1669
• Fix custom client registration by @olgavrou in #1653
• Update Transparency FAQs by @gagb in #1672
• Update agent_chat.md by @FarshidShafia in #1677
• Update notebook contrib guidance, update a few notebooks for site by @jackgerrits in #1651
• Validate the OpenAI API key format by @gunnarku in #1635
• Logging by @cheng-tan in #1146
• Validate llm_config passed to ConversableAgent (issue #1522) by @gunnarku in #1654
• do model check properly by @sonichi in #1686
• support azure assistant api by @IANTHEREAL in #1616
• Feature: Get Nested Agents in a GroupChat by @WaelKarkoub in #1636
• bug fix: logging test may fail if some config fails by @cheng-tan in #1695
• Update Azure OpenAI API version to 2024-02-15-preview by @Xtrah in #1692

New Contributors

• @bobbravo2 made their first contribution in #1646
• @yousonnet made their first contribution in #1642
• @FarshidShafia made their first contribution in #1677
• @gunnarku made their first contribution in #1635
• @WaelKarkoub made their first contribution in #1636

Full Changelog: v0.2.13...v0.2.14

v0.2.13

Highlights

• New extensible agent capability for long context handling. Example
• New extensible code execution interface and stateful executors. Examples upcoming.
• Documentation improvement and bug fix.
• Improvement in web surfer.

Thanks to @gagb @ekzhu @jackgerrits @mrwadams @LittleLittleCloud @olgavrou @davorrunje and all the other contributors!

What's Changed

• Add quarto install to Contribute.md by @jackgerrits in #1585
• Fix a couple of tiny issues in blog posts by @jackgerrits in #1578
• Fix typo in title by @mrwadams in #1594
• Fix: check response usage is not None by @olgavrou in #1599
• add other language drop down link to AutoGen website by @LittleLittleCloud in #1573
• Proxy PR for Long Context Capability 1513 by @gagb in #1591
• Hide table of contents on notebooks page by @jackgerrits in #1600
• Code executors by @ekzhu in #1405
• Refactoring web surfer to use function decorators by @davorrunje in #1435
• add long context handling notebook by @sonichi in #1618

New Contributors

• @mrwadams made their first contribution in #1594

Full Changelog: v0.2.12...v0.2.13

v0.2.12

Highlights

• Introducing AutoAnny: A New Discord Bot Built with AutoGen
• Enhancing Multiple sequential chats, a newly introduced interface that allows one to program a sequence of chats and carry previous chat results to new chats
• Allowing SocietyOfMind agents to work with function
• Documentation and usability improvements

Thanks to @gagb @afourney @jackgerrits @sonichi @joshkyh @amihalik and all the other contributors.

What's Changed

• Allow SocietyOfMind agent to work with function calling and tools. by @afourney in #1569
• Error handling in getting LLM-based summary by @qingyun-wu in #1567
• Quick Fix on the Google Colab Link by @joshkyh in #1571
• Update devcontainer.json with copilot extension by @gagb in #1556
• Introducing AutoAnny: A New Discord Bot Built with AutoGen by @gagb in #1501
• Using a more robust "reflection_with_llm" summary method by @qingyun-wu in #1575
• Expose filter_config function by @jackgerrits in #1582
• Adding max_tokens to autogenstudio LLMConfig by @amihalik in #1574
• Add mypy config to pyproject.toml by @jackgerrits in #1543
• Add more descriptive error message in the presence of OpenAI API timeouts by @jackgerrits in #1577
• Update base container for website by @gagb in #1584
• avoid access private variable by @sonichi in #1592

New Contributors

• @amihalik made their first contribution in #1574

Full Changelog: v0.2.11...v0.2.12

v0.2.11

Highlights

Two major upgrades of conversation programming (experimental features under active development):

• Finite state machine (FSM) based group chat: allow more fine-grained control of speaker order transition in group chat.
• Multiple sequential chats: program a sequence of chats and carry previous chat results to new chats.

AutoGen Studio updates:

• Workflow export.
• Skill editing.
• CSV support.
• Windows testing.

Other improvements:

• Allow timeout for code execution in windows.
• Make every agent receive the termination message in group chat.
• Improvement in documentation.
• Fix an import issue.

Thanks to @joshkyh @freedeaths @qingyun-wu @victordibia @pretherford @ekzhu @jackgerrits @samuelogboye @Xtrah @olgavrou @thinkall @abhaymathur21 and all the other contributors!

What's Changed

• Disable default code execution on society_of_mind and web_surfer. by @afourney in #1516
• Add notebooks section on website by @jackgerrits in #1495
• Update pyproject.toml for Poetry mismatch by @agamm in #1454
• Update tests for code_utils and use ThreadPoolExecutor for code execution in windows by @ekzhu in #1472
• remove useless code in groupchat; simplify the handling of termination msg by @sonichi in #1532
• Fix example for LLM configuration docs by @jackgerrits in #1528
• fix(path): Update Installation.md -Update Dockerfile path for Autogen base image build by @samuelogboye in #1425
• Update outdated link in README.md by @Xtrah in #1541
• Fix formatting of notebook contributing.md by @jackgerrits in #1540
• Custom Model Client docs follow-up by @olgavrou in #1545
• Fix docstrings of a_initiate_chat by @thinkall in #1539
• Autogenstudio Updates [CSV support, Workflow Export, Skill Editing, Windows Testing ] by @victordibia in #1475
• Graph group chat by @joshkyh in #857
• Update Docker.md by @abhaymathur21 in #1554
• patch to graph groupchat by @sonichi in #1555
• Import oai.client directly in ConversableAgent by @AkashKulkarni4444 in #1557
• New initiate_chats Interface for Managing Dependent Chats in ConversableAgent by @qingyun-wu in #1402

New Contributors

• @agamm made their first contribution in #1454
• @samuelogboye made their first contribution in #1425
• @Xtrah made their first contribution in #1541
• @abhaymathur21 made their first contribution in #1554
• @AkashKulkarni4444 made their first contribution in #1557

Full Changelog: v0.2.10...v0.2.11

v0.2.10

Breaking change

Change code_execution_config default in ConversableAgent to False to match the default value change of last_n_messages.

Highlights

• Custom model client for extensibility of the inference mechanism
• SocietyOfMindAgent: demonstrating using nested chat to compose a more capable single agent based on multi-agent chat
• Improvement of tool call and function call and GPTAssistantAgent
• Documentation improvement for function call, gallery, FAQ, notebooks etc.

Thanks to @olgavrou @afourney @davorrunje @jtrugman @ekzhu @namanbarkiya @maxim-saplin @jackgerrits @Yanni8 @victordibia @eltociear @pmalarme and all the other contributors!

What's Changed

• Update function call doc with example of not using decorator syntax. by @ekzhu in #1441
• Improve docs/gallery card component by @namanbarkiya in #1445
• FAQ, working with LLM endpoints and explaining why config-list is a list by @maxim-saplin in #1451
• fixed wrong doc link by @Yanni8 in #1449
• Removed "Tool Call Id" from main content string. by @afourney in #1471
• docs: initial Jupyter support for website docs, move config notebook by @jackgerrits in #1448
• Adds a SocietyOfMindAgent that presents as a single agent, but runs GroupChat as an inner-monologue by @afourney in #890
• Function calling upgrade by @davorrunje in #1443
• fix: unit test should not call private function by @olgavrou in #1494
• Bump autogenbench version. by @afourney in #1485
• update readme to add note on required Quarto Version. Update readme t… by @victordibia in #1493
• Update README.md by @eltociear in #1491
• Implement Overwrite Tools Functionality in GPTAssistantAgent by @jtrugman in #1434
• Update Contribute.md, #1502 by @victordibia in #1508
• Fix image print for auto feedback from code notebook by @pmalarme in #1389
• deprecate using None for code_execution_config by @jackgerrits in #1506
• Added new models to token_count_utils by @afourney in #1511
• change code_execution_config default by @sonichi in #1518
• Fix tests for GPT assistant by @davorrunje in #1505
• fix broken links from moving oai utils notebook by @jackgerrits in #1497
• Custom Model Client support by @olgavrou in #1345

New Contributors

• @namanbarkiya made their first contribution in #1445
• @jtrugman made their first contribution in #1434
• @pmalarme made their first contribution in #1389

Full Changelog: v0.2.9...v0.2.10