Don't let complexity drive your AI projects.

Drive your AI projects with innovation, not complexity.

New Standards in AI-Native App Development

‍Open Innovation Unleashed

TaskingAI brings Firebase's simplicity to AI-native app development. Start your project by selecting an LLM model, build a responsive assistant supported by stateful APIs, and enhance its capabilities with managed memory, tool integrations, and augmented generation system.

Get Started

GitHub Repo

Next-Gen Solutions for
LLM Application Development

TaskingAI offers a user-friendly UI, in a framework with separated front-end and back-end.
Ideal to all developers for efficient, flexible LLM app development.

Sleek and Intuitive Console

Integrated with tool invocation and knowledge retrieval.

Comprehensive Client SDK

REST API, Python, TypeScript, and more in the future.

Adaptable Model Integration

Integrate with leading LLM providers to ensure broad compatibility.

Quick Start

Kickstart your LLM application quickly with TaskingAI's extensive resources.

APIs for Developers:
Simplify Your Workflow

Experience the ease of building stateful assistants and executing asynchronous operations with our developer-friendly APIs.

chat_completion_result = taskingai.inference.chat_completion( model_id="YOUR_MODEL_ID", messages=[ SystemMessage("You are a professional assistant."), UserMessage("Hi"), ] )

taskingai.assistant.create_message( assistant_id=assistant.assistant_id, chat_id=chat.chat_id, text="What is the weather like?" ) taskingai.assistant.generate_message( assistant_id=assistant.assistant_id, chat_id=chat.chat_id, system_prompt_varibales={ "user_location": "New York" } )

collection = taskingai.retrieval.create_collection( embedding_model_id=embedding_model_id, capacity=1000, ) record: Record = taskingai.retrieval.create_record( collection_id=collection.collection_id, content="Machine learning is a subfield of artificial intelligence...", text_splitter=TokenTextSplitter(chunk_size=200, overlap_size=20), )

assistant = taskingai.assistant.create_assistant( model_id="YOUR_MODEL_ID", memory=AssistantNaiveMemory(), tools=[AssistantTool( type=AssistantToolType.ACTION, id=action_id, )], retrievals=[AssistantRetrieval( type=AssistantRetrievalType.COLLECTION, id=collection_id, )], )

chunks = taskingai.retrieval.query_chunks( collection_id="YOUR_COLLECTION_ID", query_text="Machine learning is ...", top_k=3 )

chat_completion = await a_chat_completion( model_id="YOUR_MODEL_ID", messages=[ SystemMessage("You are a professional assistant."), UserMessage("Count from 1 to 50"), ] )

TaskingAI's Edge Over Others

TaskingAI excels with unlimited collaboration, open-source adaptability, and a scalable, API-centric architecture for seamless AI integration and enhanced privacy.

LangChain

OpenAI Assistant API

TaskingAI

Project Members

Unlimited

Workspace roles

Admin only

Admin & Collaborators

Ecosystem

Open Source

Closed Source

Open Source

Retrieval System

Rely on 3rd-Party

Assistant Integrated Only

Decoupled

Tool Integration

Multiple LLM Providers

Custom Memory

API-Oriented

Multi-Tenant Project

Local Deployment

Roadmap: A Path to Open Innovation

Discover TaskingAI's roadmap, a strategic plan guiding us towards open-source expansion and feature-rich enhancements in AI technology.

01

Initial Reveal

TaskingAI’s official debut, showcasing its state-of-the-art AI-native app development platform.

02

Open Accessibility

TaskingAI’s client SDK and REST API Reference released to the open-source community.

03

Full-stack Deployment

Incorporating Docker enables straightforward deployment of the full-stack TaskingAI, with 100% of the source code open-sourced.

04

App by TaskingAI

Unveiling open-source applications developed by TaskingAI, with community engagement.

05

Upgraded Features

We continuously integrate new features and innovations, shaped by community feedback and the latest in AI technology.

New Standards in AI-Native App Development

‍Open Innovation Unleashed