FastGPT

FastGPT

English | Simplified Chinese | Japanese

FastGPT is a knowledge base question and answer system based on the LLM large language model, providing out-of-the-box data processing, model call and other capabilities. At the same time, workflow arrangement can be performed through Flow visualization, thereby achieving complex question-and-answer scenarios!

fastgpt.mp4

? Online use

• ? International version: tryfastgpt.ai

RoadMap

1 Application orchestration capability

• Dialogue workflow, plug-in workflow
• Tool calls
• Code sandbox
• Loop call
• User selection
• Form input

2 Knowledge Base Ability

• Multi-store multiplexing, mixed use
• chunk record modification and deletion
• Support manual input, direct segmentation, QA split import
• Support txt, md, html, pdf, docx, pptx, csv, xlsx (if you need more PR file loaders)
• Supports url reading and CSV batch import
• Mixed Search & Reorder
• Custom file reading service
• Custom chunking service

3 Application debugging capabilities

• Knowledge Base Single-point Search Test
• Feedback references during conversations and can be modified and deleted
• Full context rendering
• Full module intermediate value rendering
• Advanced orchestration DeBug mode

4 OpenAPI interface

• completions interface (chat mode alignment GPT interface)
• Knowledge Base CRUD
• Conversation CRUD

5 Operational capabilities

• Login-free sharing window
• Iframe one-click embed
• Chat window embed supports custom Icon, default opening, dragging and dropping functions
• Unified review of dialogue records and annotate the data

6 Others

• Support voice input and output (can be configured for voice input and voice answers)
• Fuzzy input prompts
• Template Market

?‍ Development

Project technology stack: NextJs + TS + ChakraUI + MongoDB + PostgreSQL (PG Vector Plugin)/Milvus

• ⚡ Quick deployment

  Using Sealos services, no need for procurement servers, no domain names, supports high concurrency and dynamic scaling, and database applications use kubeblocks databases, which far exceeds the simple Docker container deployment in terms of IO performance.

  Click to view the Sealos one-click deployment of FastGPT tutorial

• Quickly start local development
• Deploy FastGPT
• System configuration file description
• Multi-model configuration
• Version update/upgrade introduction
• OpenAPI API Documentation
• Detailed explanation of the knowledge base structure

?️ Community Exchange Group

Scan the QR code to join the Feishu topic group (newly opened, gradually abandoning the WeChat group):

? Related items

• Laf: 3 minutes quick access to three-party applications
• Sealos: Rapid deployment of cluster applications
• One API: Multi-model management, supports Azure, Wenxin Yiyan, etc.
• TuShan: Build a backend management system in 5 minutes

? other

• Nanny-level FastGPT Tutorial
• Access to Feishu
• Access to Qiwei

? Third-party ecology

• COW Personal WeChat/Qiwei Robot

? Participate in contributions

We highly welcome contributions in all forms. If you are interested in contributing code, check out our GitHub Issues to show us your whims.

? Star History

Usage Protocol

This repository follows the FastGPT Open Source License open source protocol.

1. It is allowed to be commercially available as a backend service, but SaaS services are not allowed.
2. No commercial authorization is required to retain relevant copyright information for any form of commercial use.
3. Please check out the FastGPT Open Source License for complete
4. Contact information: [email protected], click to view the commercial version pricing strategy