#1919th of 35 in Agent platforms
bisheng
Enterprise LLM app platform with visual workflows, knowledge spaces and a task agent
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 12.1k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Aug 2026
- Language
- Python
11,161 on 31 Mar 202612,068 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
BISHENG is an LLM application platform aimed at enterprises. It combines a visual workflow editor (loops, parallel branches, batch runs, human-in-the-loop pauses), a team knowledge base with file-level permissions and cited AI Q&A, and a Task Mode agent that plans work, runs sub-agents in parallel and produces PPT, Excel or Word output. It also ships department management, approvals, audit logs and an open API.
Who it is for: Enterprise teams building internal LLM apps, knowledge bases and agents
Strengths
- Workflow editor supports loops, parallel branches and batch runs in one flow with chat
- RBAC plus ReBAC permissions down to single files, with department and user-group grants
- Includes SSO/LDAP, approvals, audit logs and an open API with service accounts
- Document parsing models (OCR, tables, layout, seals) can be deployed privately for free
Weaknesses
- Heavy install: Elasticsearch, Milvus and Onlyoffice by default; 16 GB RAM minimum
- Dashboards and DSH Desktop enterprise access are commercial edition only
- Docs and community channels are mostly Chinese (Feishu wiki, WeChat group)
- README lists no supported model providers or GPU requirements
What it needs
- RAM ≥ 16 GB
- Compose
- Needs Docker, Docker Compose, Elasticsearch, Milvus, Onlyoffice, MySQL or Dameng DM8
- port 3001
Also in Agent platforms
See all 35| Rank | Project | Score |
|---|---|---|
| 1 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 81 out of 100 |
| 2 | daily_stock_analysisDaily LLM stock analysis for A-share, HK, US and other markets, pushed to chat apps | 80 out of 100 |
| 3 | deer-flowAgent harness that runs sub-agents, memory and sandboxed skills for long tasks | 77 out of 100 |
| #4 | vibe-tradingFinance research agent for backtests, market data and broker connectors | 77 out of 100 |
| #5 | MulticaIssue board where AI coding agents take assignments like teammates | 76 out of 100 |