#9494th of 102 in Agent platforms
cultivation-world-simulator
Xianxia world simulator where every NPC is driven by an LLM
- Stars
- 2.2k
- License
- custom license
- Last commit
- Aug 2026
- Last release
- Aug 2026
- Language
- Python
1,514 on 25 Apr 20262,150 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
A simulation game in which each cultivator is an independent LLM agent with personality, memory, relationships and goals, bounded by a rule system covering realms, sects, techniques, pills and events. The player acts as an observer who can intervene. It runs from source (FastAPI backend, Vue/PixiJS frontend), via docker-compose, or as an Epic Games Store desktop build, and exposes a query/command HTTP API for external agents.
Who it is for: Players and developers who want an LLM-driven emergent-narrative sim
Strengths
- Per-NPC LLM agents with short and long-term memory, goals and relationships, bounded by game rules
- Model presets for DeepSeek, MiniMax and Ollama, so local models work
- Versioned /api/v1 query and command endpoints let external agents observe and intervene
- Three install paths: source, docker-compose, or Epic Games Store desktop build
Weaknesses
- Docker deployment is labeled untested in the README
- Many concurrent LLM agents imply heavy API usage; cost and throughput are not documented
- Mobile UI is not fully adapted
- Main README is in Chinese; license is not identified (NOASSERTION)
What it needs
- GPU optional
- Docker + Compose
- Needs Python 3.10+, Node.js 18+, LLM provider or Ollama
- Models: DeepSeek, MiniMax, Ollama, custom LLM API
- port 8123
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Agent platforms
See all 102| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, served as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |