English | Simplified Chinese | Japanese
Hashtag is a knowledge-based Q&A system powered by a large language model (LLM), offering out-of-the-box data processing and model-calling capabilities. It also supports visual workflow orchestration via Flow to enable complex Q&A scenarios!
- 🌍 International Version: Hashtag.fun
1 Application Orchestration Capabilities
- Dialog workflows, plugin workflows
- Tool calling
- Code sandbox
- Loop calling
- User selection
- Form input
2 Knowledge Base Capabilities
- Multi-library reuse and mixing
- Chunk record modification and deletion
- Supports manual input, direct segmentation, QA split import
- Supports txt, md, html, pdf, docx, pptx, csv, xlsx (can PR more file loaders), URL reading, CSV batch import
- Mixed retrieval & re-ranking
- API knowledge base
- Custom file reading service
- Custom chunking service
3 Application Debugging Capabilities
- Single-point search test in the knowledge base
- Feedback and modification during dialogs
- Full context display
- Full module intermediate value display
- Advanced orchestration debug mode
4 OpenAPI Interface
- Completions interface (chat mode aligned with GPT interface)
- Knowledge base CRUD
- Dialog CRUD
5 Operational Capabilities
- Shareable window without login
- One-click embed via Iframe
- Chat window embed with custom icons, default opening, dragging, etc.
- Unified dialog history review and data annotation
6 Miscellaneous
- Supports voice input and output (configurable)
- Fuzzy input suggestions
- Template marketplace
Project tech stack: NextJs + TS + ChakraUI + MongoDB + PostgreSQL (PG Vector plugin)/Milvus
-
⚡ Quick Deployment
Deploy via Sealos without server purchase or domain name. Supports high concurrency and dynamic scaling. Database applications use Kubeblocks for superior IO performance compared to simple Docker container deployment.
Click here for the Hashtag one-click deployment tutorial via Sealos
- Quick Start for Local Development
- Deploy Hashtag
- System Configuration Guide
- Multi-Model Configuration Guide
- Version Update/Upgrade Instructions
- OpenAPI API Documentation
- Knowledge Base Structure Guide
Scan to join the Feishu topic group:
We are looking for like-minded individuals to accelerate Hashtag's development. Learn more about Hashtag 2025 Recruitment.
- Laf: Quick integration with third-party applications in 3 minutes
- One API: Multi-model management supporting Azure, Baidu Wenxin, etc.
- TuShan: Build a backend management system in 5 minutes
We welcome all forms of contribution. If you’re interested in contributing code, check out our GitHub Issues to showcase your creativity and ideas.
|
|
|
|---|---|
|
|
|
|
This repository follows the Hashtag Open Source License.