Bisheng
Bisheng is an open-source LLM application DevOps platform designed for enterprise scenarios, enabling the development, deployment, and management of complex AI applications. It solves the problem of building production-grade LLM applications with features like workflow orchestration, human-in-the-loop, and high-precision document parsing. It is for developers and organizations looking to implement enterprise-level AI solutions.
✨ Key features
- Lingsight agent with expert-level taste via AGL framework
- Unique workflow orchestration with loops, parallelism, batch processing
- Human-in-the-loop intervention during workflow execution
- Enterprise-grade features: RBAC, SSO/LDAP, security review
- High-precision document parsing models for text, tables, seals
- Community for sharing best practices and application cases
🎯 Use cases
- Document review and fixed-layout report generation
- Multi-agent collaboration and policy update comparison
- Support ticket and customer service assistance
- Meeting minutes generation and resume screening
- Unstructured data governance and knowledge mining
📦 Installation
🧰 Requirements: CPU >= 4 cores, RAM >= 16 GB, Docker 19.03.9+, Docker Compose 1.25.1+; recommended 18 cores, 48 GB RAM.
git clone https://github.com/dataelement/bisheng.git
cd bisheng/dockerwget https://github.com/dataelement/bisheng/archive/refs/heads/main.zip
unzip main.zip && cd bisheng-main/dockerStart with:
docker compose -f docker-compose.yml -p bisheng up -d
Access http://IP:3001 and register the first user as admin.
🚀 Usage
After starting, open http://IP:3001 in a browser, register a new user (first user becomes admin), then log in to start building applications.
❓ FAQ
What are the minimum hardware requirements?
CPU >= 4 virtual cores, RAM >= 16 GB, Docker 19.03.9+, Docker Compose 1.25.1+. Recommended: 18 cores, 48 GB RAM.
How do I access the platform after installation?
After running docker compose up, access http://IP:3001 in a browser and register the first user, who becomes the system admin.
Does Bisheng support document parsing?
Yes, it includes high-precision document parsing models for printed text, handwriting, tables, layout, and seals, which can be deployed privately for free.
What enterprise features are included?
It includes security review, RBAC, user group management, traffic control, SSO/LDAP, vulnerability scanning, high availability deployment, monitoring, and statistics.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.