v0.3.0 🚀

TaskingAI Updates: Version 0.3.0

What's New

1. Markdown content rendering and image uploads are now supported in Playground.
   

2. We introduce an OpenAI-compatible chat completion API, allowing users to seamlessly create Language Learning Model (LLM) based tasks or agent-based invocations.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_TASKINGAI_API_KEY",
    base_url="https://oapi.tasking.ai/v1",
)
response = client.chat.completions.create(
        model="YOUR_TASKINGAI_MODEL_ID", # or an assistant ID
        messages=[
            {"role": "user", "content": "Hello, how are you?"},
        ]
    )

print(response)

3. Inference and Plugin service is now open source!

Key Fixes and Improvements

• Error Log Updates:
  • Updated error log handling for Google Gemini when using function in v1 #106.
• Build and Deployment Enhancements:
  • Updated server version in Docker Compose #111.
  • Bumped aiohttp to version 3.9.4 #109 and gunicorn to version 22.0.0 in the backend #108.
  • Added NLTK data to server Docker #158.
• Inference and Model Management:
  • Updated inference version #116 and fixed wildcard model inference issues #107.
  • Added rerank model #144 and associated tag #162.
  • Fixed issues related to creating wildcard models during collection creation #119.
• User Interface and Experience:
  • Introduced a new separator text splitter #115 and a name column to chat #118.
  • Updated navigation bar UI structure #128 and added project home #124.
  • Added chat search bar #129, playground markdown content support #132, and image upload in the playground #133.
  • Improved plugin refresh logic after creation #166 and deletion #167.
• Validation and Configuration:
  • Implemented validation for action credentials during creation and updating #113.
  • Added a score threshold to retrieval config #117.
  • Added a checkbox for playground model configuration #126.
• Sample Codes and Quick Access:
  • Added quick access to example codes #130 and sample code API #136.
  • Added a flag to determine whether to include file category #160.

New Features

• Frontend and Playground Enhancements:
  • Added a stateless assistant invocation feature #137.
  • Added support for file and web records #131, and local image mount access #140.
  • Added assistant and collection UI API #138.
  • Made the inference service open-source #127.
  • Made the plugin service open-source #155.
• OpenAI Compatibility:
  • Introduced OpenAI compatible API #139.

Testing and Quality Assurance

• Backend Testing:
  • Added test cases for backend #125.
  • Updated inference test cases #154 and ignored unavailable test cases #146.
  • Added AWS storage test #165.
• Error Handling and Fixes:
  • Fixed typos and spelling errors #122 and #142.
  • Fixed API routes #141 and handled empty record content #152.
  • Fixed chunk table pagination #169.
  • Fixed local file original name #153.
  • Fixed issues with deleting files in local storage #161.

Miscellaneous

• Chore and Dependency Updates:
  • Updated version #148 and removed unnecessary packages #164.
  • Bumped requests from 2.31.0 to 2.32.2 in the inference module #145.
  • Updated plugin dependencies #163.
  • Added frontend Docker build test #143.

What's Changed

• fix: update error log of google gemini when use function in v1 by @SimsonW in #106
• build: update server version in docker-compose by @jameszyao in #111
• chore: update inference version by @SimsonW in #116
• fix: fix wildcard model inference by @jameszyao in #107
• fix: fixed the issue where cannot create wildcard models during collection creation by @DynamesC in #119
• feat: validate action credentials when creating and updating by @DynamesC in #113
• feat: add separator text splitter by @jameszyao in #115
• feat: add name column to chat by @jameszyao in #118
• feat: add score threshold to retrieval config by @jameszyao in #117
• feat: frontend add project home by @taskingaiwww in #124
• feat: add checkbox for playground model configuration by @taskingaiwww in #126
• test: add test for backend by @taskingaijc in #125
• fix: fix typos by @taskingaiwww in #122
• feat: update the navigation bar UI structure by @taskingaiwww in #128
• build(deps): bump aiohttp from 3.9.3 to 3.9.4 in /backend by @dependabot in #109
• feat: add chat search bar in the assistant playground by @taskingaiwww in #129
• feat: add quick access to example codes by @taskingaiwww in #130
• feat: add playground markdown content support by @taskingaiwww in #132
• feat: image upload in the playground by @taskingaiwww in #133
• feat: file and web records by @Dttbd in #131
• feat: add sample code API by @taskingaijc in #136
• feat: stateless assistant invocation by @jameszyao in #137
• feat: add assistant and collection ui api by @Dttbd in #138
• build(deps): bump gunicorn from 21.2.0 to 22.0.0 in /backend by @dependabot in #108
• fix: fix spelling errors by @Dttbd in #142
• fix: fix api routes by @taskingaiwww in #141
• feat: OpenAI compatible API by @SimsonW in #139
• ci: add frontend docker build test by @jameszyao in #143
• feat: inference service is now open-source by @SimsonW in #127
• build(deps): bump requests from 2.31.0 to 2.32.2 in /inference by @dependabot in #145
• fix: ignore unavailable test case by @SimsonW in #146
• fix: ignore unavailable provider test case by @SimsonW in #147
• chore: update version by @Dttbd in #148
• feat: add web and file record types by @taskingaiwww in #149
• fix: fix typos by @taskingaiwww in #151
• feat: add local image mount access by @Dttbd in #140
• fix: handle empty record content by @jameszyao in #152
• test: update inference test case by @SimsonW ...

v0.2.2

What's Changed

• feat: allow model config update by @jameszyao in #91
• fix: message generation stability by @jameszyao in #90
• fix: read num_chunk before record update by @jameszyao in #95
• feat: add tool result size check by @jameszyao in #92
• feat: add collection capacity check by @jameszyao in #100
• feat: supports array as input of plugin by @DynamesC in #101
• feat: update model module UI by @taskingaiwww in #103
• feat: add chat completion playground frontend by @taskingaiwww in #103
• feat: add reset credentials button in edit action drawer by @taskingaiwww in #103
• chore: update inference version by @SimsonW in #102
• build: update docker compose to v0.2.2 by @jameszyao in #104

Full Changelog: v0.2.0...v0.2.2

v0.2.0

What's Changed

• feat: add "Generate and Send" button in assistant playground by @jameszyao in #38
• build(deps): bump python-multipart from 0.0.6 to 0.0.7 in /backend by @dependabot in #36
• Fix issue with settingModal undefine warnings by @jishen027 in #39
• fix(frontend): correct typo of Fucntion -> Function by @bufferoverflow in #42
• feat: add plugin module by @jameszyao in #43
• Fix/fix collection update api by @jameszyao in #45
• feat: add plugin frontend module by @taskingaiwww in #46
• fix: update equal filter in list api route by @jameszyao in #47
• Build/update docker compose 0.2.0 by @jameszyao in #48
• doc: add integration introduction to README.md by @jameszyao in #44
• fix: update model type key when verify model credentials by @jameszyao in #50
• fix: frontend add model properties when creating wildcard model by @taskingaiwww in #52
• build: update docker-compose.yml by @jameszyao in #53
• fix: mark model name as required by @taskingaiwww in #55
• fix: fix stream message generation in debug model by @jameszyao in #54
• build: update docker compose to v0.2.0 by @DynamesC in #56
• fix: fix content value by @taskingaiwww in #57

New Contributors

• @dependabot made their first contribution in #36
• @jishen027 made their first contribution in #39
• @bufferoverflow made their first contribution in #42
• @taskingaiwww made their first contribution in #46

Full Changelog: v0.1.3...v0.2.0

v0.1.3

What's Changed

• fix: check input properties when create model by @jameszyao in #32
• feat: add tooltip to record and chunk content by @taskingaiwww in #32
• fix: fix not found page UI by @taskingaiwww in #32
• build: update docker-compose.yml by @jameszyao in #32

Full Changelog: v0.1.2...v0.1.3

v0.1.2

What's Changed

• fix version for docker-compose.yml by @superiorlu in #21
• feat: add custom model properties by @jameszyao in #27
• fix: use model properties in retrieval collection by @jameszyao in #27
• feat: add retrieval chunks by @jameszyao in #27
• fix: fix provider_name by @jameszyao in #27
• feat: hide empty properties by @jameszyao in #27
• build: update docker-compose.yml by @jameszyao in #27

New Contributors

• @superiorlu made their first contribution in #21

Full Changelog: v0.1.1...v0.1.2

v0.1.1

What's Changed

• docs: restructure readme and add demo video by @DynamesC in #9
• chore: update code of conduct by @SimsonW in #12
• feat: update action schema by @jameszyao in #16
• doc: update README.md by @jameszyao in #17
• feat: support custom host model with OpenAI API format @jameszyao
• feat: add streaming chat completion support for models from Mistrial AI @jameszyao

New Contributors

• @eltociear made their first contribution in #13

Full Changelog: v0.1.0...v0.1.1

v0.1.0

What's Changed

• chore: update README.md by @jameszyao in #2
• docs: update README.md by @jameszyao in #3
• build: update docker-compose.yml by @jameszyao in #4

Full Changelog: https://github.com/TaskingAI/TaskingAI/commits/v0.1.0