v0.2.2

What's Changed

• Add result returning to batch processing by @djwu563 in #142
• Add max_tasks to start_batch_processor in ProgrammaticClient by @djwu563 in #143
• Add custom tasks and integrate with the AaaS platform by @djwu563 in #145
• update docs by @djwu563 in #144
• Develop/v0.2.2: support passing variables in the prompt template through config file, update llm.md by @XeonHis in #149
• add Aaas device by @Silentharry94 in #151
• Update: Add inputs for DoWhileTask by @P3ngLiu in #152
• Add domain validation for Aaas by @djwu563 in #153
• Add worker.id to assign it to task_definition_name by @djwu563 in #154
• Feature/v0.2.2/prevent_module_duplication by @panregedit in #157
• Feature/v0.2.2/aaas by @Silentharry94 in #161
• opt Aaas message type and event value by @Silentharry94 in #164
• Update the "Quick Start" section in the README by @djwu563 in #166
• Add chatgpt parameter by @noname-zh in #165
• Develop/v0.2.2: update tavily search doc, fix minor bug in video understanding stage by @XeonHis in #158
• add ollama doc by @yileld in #171
• Feature/v0.2.2/PoT by @P3ngLiu in #160
• add CoT operator and example by @KingSan666888 in #150
• Self consist cot workflow by @JulienZe in #159
• add React and React_pro by @lijingcheng2021 in #169
• opt workflow_instance_id type add task execute timeout by @Silentharry94 in #170
• Develop/v0.2.2: Support token usage of each running, isolated by workflow_instance_id by @XeonHis in #172
• opt func_timeout by @Silentharry94 in #174
• readme add input and output by @JulienZe in #176
• improve the code execution by @P3ngLiu in #177
• Refactor: Remove init method and rewrite search functionality for improved modularity by @lijingcheng2021 in #180
• Add support for setting a domain for workers by @djwu563 in #179
• opt conversation_info by @Silentharry94 in #178

New Contributors

• @noname-zh made their first contribution in #165
• @yileld made their first contribution in #171
• @JulienZe made their first contribution in #159
• @lijingcheng2021 made their first contribution in #169

Full Changelog: v0.2.1...v0.2.2

v0.2.1

1. New debug mode, features include:
   • More detailed log information
   • Stops all running instances of the same workflow when a workflow is started
   • Disables error retries
2. New Tavily search tool
3. New device:
   • WebpageClient provides a Gradio-based web UI
   • ProgrammaticClient offers a concise programming interface, allowing you to call workflows as Python functions
4. Use the conductor's logging system instead of the local logging system.
5. New in-memory STM, enabling STM functionality without relying on external components
6. Optimized large language model invocation methods, especially for multimodal invocation.
7. Optimized app backend code

v0.2.0

What's Changed

• refactor the framework by @panregedit in #26
• Refactor Request Handling and Enhance Error Logging in general.py by @Mefisto04 in #28
• Add rescue node by @panregedit in #29
• fix import error by @panregedit in #30
• Feature/v0.2.0/conductor by @panregedit in #34
• add redis stream listener and app callback by @djwu563 in #36
• Refactor STM and fix errors by @P3ngLiu in #37
• Feature/v0.2.0/app_client by @panregedit in #38
• Dev/v0.2.0 by @panregedit in #39
• Add MilvusLTM: LTM with Milvus Vector DB by @kyusonglee in #40
• Bugfix: fix forktask compile error by @XeonHis in #41
• fix stm_redis bug, change decode_responses=False to RedisConnector by @P3ngLiu in #43
• Feature/v0.2.0/opt_container by @panregedit in #45
• fix app client input and callback bug by @djwu563 in #44
• Feature/v0.2.0/build_worker by @panregedit in #46
• Hotfix/v0.2.0/container_wrong_input by @panregedit in #47
• fix input and callback bug by @djwu563 in #48
• Optimize the client module and remove unnecessary packages by @djwu563 in #49
• Hotfix/v0.2.0/container_load by @panregedit in #50
• Feature/v0.2.0/worker_supports_setting_concurrency by @panregedit in #51
• Feature/v0.2.0/DnC: refactor Divide-and-Conquer feature to conductor compatible implementation by @XeonHis in #52
• Feature/v0.2.0/share_workflowId by @panregedit in #53
• Feature/v0.2.0/opt_container_compile by @panregedit in #54
• update ltm and stm, fix bug by @P3ngLiu in #56
• Feature/v0.2.0/dict_stm by @panregedit in #57
• Add support for multi-line input to DefaultClient and enable the clie… by @djwu563 in #58
• Feature/v0.2.0/opt_conductor_config by @panregedit in #59
• Hotfix/v0.2.0/merge_compile_container by @panregedit in #60
• Hotfix/v0.2.0/conductor_config by @panregedit in #61
• Feature/v0.2.0/DnC: refactor AgentTask, callback function update, compatible with new stm and conductor configurations by @XeonHis in #62
• add examples step1_simpleVQA by @P3ngLiu in #63
• Fix the bug in the CLI input and adjust the code in run.py by @djwu563 in #64
• Feature/v0.2.0/DnC: Be compatible with new run_app and run_cli entrypoint by @XeonHis in #65
• Feature/v0.2.0/opt_client_import by @panregedit in #66
• add examples by @P3ngLiu in #67
• Bug fix: reset redis_stream_client password to default, fix register stm error by @XeonHis in #68
• Bug fix: reset redis_stream_client password to default, fix register stm error by @XeonHis in #69
• update readme for step1 simple vqa by @P3ngLiu in #70
• Develop/v0.2.0 by @djwu563 in #71
• add readme for examples step1 and step2 by @P3ngLiu in #72
• update readme for examples by @P3ngLiu in #74
• Feature/v0.2.0/add_docs by @panregedit in #75
• Feature/v0.2.0/add_docs by @panregedit in #76
• update websearch.yml, and update README by @djwu563 in #77
• bugfix: give init input prompt by @XeonHis in #79
• update ltm to skip create collection in init by @P3ngLiu in #78
• add app docs by @djwu563 in #80
• update readme and app docs by @djwu563 in #81
• Feature/v0.2.0/add_docs by @panregedit in #82
• Feature/v0.2.0/video understanding: Refactor single round video understanding Q&A to conductor-based implementation by @XeonHis in #83
• Feature/v0.2.0/app_deploy by @panregedit in #84
• fix deploy issue by @panregedit in #85
• fix app deploy by @panregedit in #86
• update readme for git lfs pull by @P3ngLiu in #87
• fix om-app-agent deploy by @panregedit in #88
• Update outfit_with_ltm.md by @P3ngLiu in #89
• update app docs by @djwu563 in #90
• Feature/v0.2.0/video understanding:Optimize and update README by @XeonHis in #92
• Optimize LTM vector db create logic, update README by @XeonHis in #93
• fix incorrect reference to dnc README by @XeonHis in #94

New Contributors

• @Mefisto04 made their first contribution in #28
• @kyusonglee made their first contribution in #40

Full Changelog: v0.1.0...v0.2.0

v0.1.0

An official implementation of the OmAgent paper. Supports reproducing the DnC Loop and Video RAG mentioned in the paper.

What's Changed

• docs: add Japanese README by @eltociear in #12
• docs :update README_FR.md by @Innovatorcloudy in #15
• Update README_ZH.md by @kushal34712 in #17
• Update README.md by @kushal34712 in #20
• [DOC] fixed some typos in README.md by @Smoothengineer in #21
• Updated README_FR.md by @Naman9761 in #22

New Contributors

• @eltociear made their first contribution in #12
• @Innovatorcloudy made their first contribution in #15
• @kushal34712 made their first contribution in #17
• @Smoothengineer made their first contribution in #21

Full Changelog: https://github.com/om-ai-lab/OmAgent/commits/v1.0.0