1. Project Overview

FastGPT

FastGPT is a knowledge base question-answering system built on LLMs. It provides ready-to-use capabilities for data processing, model invocation, and more. It also supports visual workflow orchestration through Flow, enabling complex question-answering scenarios!

2. Open Source License

Licensed under Apache-2.0

3. Screenshots

FastGPT Capabilities

1. AI Customer Service

Train AI models by importing documents or existing question-and-answer pairs, enabling them to answer questions based on your documentation.

2. Automatic Data Preprocessing

Provides multiple data import methods, including manual input, direct segmentation, automatic LLM processing, and CSV, catering to both precise and rapid training scenarios.

3. Workflow Orchestration

Workflow orchestration based on Flow modules helps you design more complex question-answering processes, such as querying databases, checking inventory, and booking laboratories.

4. Seamless OpenAPI Integration

FastGPT's external API interfaces are aligned with the official GPT API. You can integrate FastGPT directly into existing GPT applications by simply modifying the BaseURL and Authorization.

4. Feature Overview

FastGPT is a knowledge base question-answering system built on LLMs. It provides ready-to-use capabilities for data processing, model invocation, and more. It also supports visual workflow orchestration through Flow, enabling complex question-answering scenarios!

Get to Know FastGPT Quickly

FastGPT is an AI question-answering system based on vector embeddings and LLM models. It helps you quickly and efficiently train models using internal documents. It provides a rich set of debugging tools for data adjustment and can be integrated into your existing applications through Iframe or API.

FastGPT Features

1. Open Source Project. FastGPT follows the Apache License 2.0 open-source license. You can clone FastGPT from GitHub for secondary development and distribution. The FastGPT Community Edition retains core functionality, while the commercial edition extends it through APIs on top of the Community Edition without affecting learning and usage.

2. Unique QA Structure. Designed specifically for customer service question-answering scenarios, the QA structure improves accuracy when handling large volumes of data.

3. Visual Workflow. The Flow module displays the complete process from question input to model output, making it easier to debug and design complex workflows.

4. Unlimited Extensibility. Supports extensions based on HTTP, allowing you to quickly integrate FastGPT into existing programs without modifying its source code.

5. Convenient Debugging. Provides multiple debugging options, including search testing, citation modification, and full conversation previews.

6. Supports Multiple Models: Supports various LLM models, including GPT, Claude, and ERNIE Bot, with plans to support custom embedding models in the future.

5. Tech Stack

Development

Project Tech Stack: NextJs + TS + ChakraUI + Mongo + Postgres (Vector Plugin)

Knowledge Base Core Workflow Diagram