BISHENG is an open LLM DevOps platform for enterprise applications, providing comprehensive features including GenAI workflows, RAG, Agents, unified model management, evaluation, SFT, dataset management, and more to help organizations build next-generation AI applications。
One-Minute Overview#
BISHENG is an open-source LLM DevOps platform designed for enterprise applications, providing a complete ecosystem for developing AI solutions. It integrates core features like workflow orchestration, agents, RAG, and model management, and has been adopted by numerous industry-leading organizations and Fortune 500 companies.
Core Value: Through visual workflows and enterprise-grade features, BISHENG enables enterprises to efficiently build, deploy, and manage complex AI applications.
Quick Start#
Installation Difficulty: Medium - Requires Docker environment and multiple third-party components
git clone https://github.com/dataelement/bisheng.git
cd bisheng/docker
docker compose -f docker-compose.yml -p bisheng up -d
Is this suitable for me?
- ✅ Enterprise AI application development: BISHENG provides complete enterprise-grade features including security reviews, RBAC, user group management
- ✅ Complex workflow requirements: Its unique workflow supports loops, parallel processing, batch processing, and other complex logic
- ❌ Simple AI chatbots: For simple applications, BISHENG may be overly complex
- ❌ Resource-constrained environments: Minimum requirement of 4 CPU cores and 16GB RAM, with 18 cores and 48GB recommended
Core Capabilities#
1. Lingsight Agent - Expert-level Task Processing#
Through the AGL (Agent Guidance Language) framework, we embed domain experts' preferences, experience, and business logic into the AI, enabling the agent to exhibit "expert-level understanding" when handling tasks. Real Value: Reduces reliance on specialized expertise, allowing AI applications to solve more complex professional problems
2. Unique BISHENG Workflow#
- Independent and comprehensive application orchestration framework: Enables execution of various tasks within a single framework (while similar products rely on bot invocation or separate chatflow and workflow modules)
- Human in the loop: Allows users to intervene and provide feedback during workflow execution (including multi-turn conversations)
- Powerful: Supports loops, parallelism, batch processing, conditional logic, and free combination of all logic components Real Value: Enterprises can flexibly build complex application workflows based on actual business needs without writing extensive code
3. Enterprise Application Support#
Document review, fixed-layout report generation, multi-agent collaboration, policy update comparison, support ticket assistance, customer service assistance, meeting minutes generation, resume screening, call record analysis, and more. Real Value: Directly addresses common enterprise AI application scenarios with deeply optimized components and parameter configurations
4. Enterprise-grade Features#
Security review, RBAC, user group management, traffic control by group, SSO/LDAP, vulnerability scanning and patching, high availability deployment solutions, monitoring, statistics, and more. Real Value: Ensures security and reliability of enterprise-grade applications, meeting compliance requirements
5. High-Precision Document Parsing#
High-precision document parsing model trained on vast amounts of high-quality data accumulated over 5 years, including printed text, handwritten text, and rare character recognition models, table recognition models, layout analysis models, and seal models. Real Value: Enterprises can deploy it privately for free to process various documents and accurately extract key information
Tech Stack & Integration#
Development Languages: Python, JavaScript, HTML, CSS, TypeScript, Shell Key Dependencies: Elasticsearch, Milvus, OnlyOffice, LangChain, LangFlow, LLaMA-Factory, Unstructured Integration Method: Platform deployment
Maintenance Status#
- Development Activity: High frequency with multiple commits per week
- Recent Updates: Recent releases available
- Community Response: Active issue response and community discussions
Commercial & Licensing#
License: Apache-2.0
- ✅ Commercial Use: Allowed
- ✅ Modification: Allowed
- ⚠️ Restrictions: Attribution required
Documentation & Learning Resources#
- Documentation Quality: Comprehensive
- Official Documentation: https://github.com/dataelement/bisheng/wiki
- Sample Code: Available