Purpose
FastGPT is an AI Agent building platform. Its goal is to help users quickly build AI knowledge bases and applications by providing out-of-the-box capabilities such as data processing and model invocation. The platform supports many-to-many relationships and can perform visual workflow orchestration through Flow, thereby implementing complex application scenarios.
Main Capabilities
- Model invocation: can connect to large model services such as the OpenAI API.
- Data processing: has out-of-the-box data processing capabilities and supports knowledge base construction.
- Workflow orchestration: supports conversation workflows and plugin workflows, includes basic RPA nodes, and supports advanced orchestration and DeBug debug mode.
- Knowledge base import: supports manual input, direct segmentation, and QA split import.
- File format support: supports TXT, MD, HTML, PDF, Docx, PPTX, CSV, XLSX, and supports URL reading and CSV batch import. The repository states that a PR can be submitted if more File Loaders are needed.
- Conversation and data management: can uniformly view conversation records and annotate data.
- Supporting services: provides a model aggregation load balancing service named AI Proxy and the AI Proxy large model aggregation service; can be combined with Sealos to quickly deploy cluster applications, and mentions the SiliconCloud open-source model online experience platform.
Usage
- Docker quick start: you can enter commands in the terminal and complete the inputs according to the guidance to pull the configuration. After fully starting, you can access FastGPT through http://localhost:3000 . The default account is root and the password is 1234. If you encounter problems, you can view the complete Docker deployment tutorial.
- Cloud service version: if private deployment is not needed, you can directly use the cloud service version at fastgpt.io.
- Community self-hosted version: you can use Docker for rapid deployment, or use Sealos Cloud for one-click deployment of FastGPT.
- Commercial version: if you need more complete functionality or in-depth service support, you can choose the commercial version. In addition to the complete software, it also provides corresponding scenario implementation guidance. For details, you can submit a commercial inquiry.
Applicable Scenarios
Suitable for teams and developers who need to build AI knowledge bases, orchestrate conversation or plugin workflows, import and process multiple document formats, and need private deployment or use a hosted cloud service. It can also be used in scenarios that require model aggregation, load balancing, and cluster deployment solutions.
Restrictions and License
This repository follows the FastGPT Open Source License. It is allowed to be used directly for commercial purposes as a backend service, but it is not allowed to provide SaaS services. Without commercial authorization, any form of commercial service must retain the relevant copyright information. For complete terms, please see the FastGPT Open Source License.
Other Information
Contact: Dennis@sealos.io; the commercial version pricing strategy can be viewed through the relevant link. Other details are not provided for now.