I Built AltiVerse: Fork Decisions Into Living Simulations With 1,000 Personalities
What if you could see the real consequences of a decision before making it?
Not with abstract models or spreadsheets — but by watching full living simulations unfold.
That’s what AltiVerse is.
GitHub:
LeoTheAIDev
/
Altiverse
Fork decisions into living simulations with 1,000 personalities. Watch alternate realities diverge and explore second-order effects.
Take one choice. Play it forward as several alternate realities — each a small living world of people who move, stress out, break rules, burn out, form friendships and rivalries, and tell you how it felt.
What is AltiVerse?
AltiVerse forks a single decision — a school phone ban, a 4-day week, 8- vs 12-hour hospital shifts — into 2–4 alternate realities and runs each one forward as a small, living simulation. Up to ~1,000 people with personalities, moods, and relationships move through rooms, react to the policy, and slowly pull the worlds apart. You watch when and why the realities diverge, click any person to compare how they fare across timelines, and export a full report with a recommendation.
It is a thinking tool, not a predictor. Every number and quote comes from a deterministic engine (same seed → same world, every time), with an optional local or online…
The Idea
AltiVerse lets you take any decision (school phone ban, 4-day work week, hospital shift changes, etc.) and fork it into 2–4 alternate realities.
Each reality runs as a deterministic living simulation with up to 1,000 agents. Every agent has:
- Unique personality
- Changing moods
- Memories
- Relationships
They move through rooms, interact, react to the decision, and slowly evolve.
Key Features
- Watch the worlds evolve side-by-side
- See exactly when and why the timelines diverge
- Click any person to compare how their life turns out across all realities
- Generate detailed reports with in-character thoughts and insights
- Fully deterministic (same seed = same world every time)
Example
I forked a school phone ban into four realities.
In one timeline grades improved but anxiety spiked.
In another students rebelled and social connections collapsed.
The divergence was clear and fascinating.
How It’s Built
- Web interface for forking decisions and exploring timelines
- Optional LLM layer for natural prose in reports and thoughts
The project is fully open source.
What’s Next
I’m working on better performance, richer personalities, and more visualization tools.
What decision would you want to fork and simulate? Drop your ideas in the comments — happy to run a few and share the results.
Feedback is very welcome, especially on simulation mechanics and UX.
This version keeps it general and focused on the core idea. Add your GIFs/screenshots when publishing.
Want any tweaks?
Top comments (1)
Cool concept with the multi-agent simulation. The idea of watching agents with different personalities react to decisions is powerful. One thing I noticed when building agent systems: agents in a simulation would benefit from having distinct 'thinking modes' — some naturally more divergent while others converge quickly. I put together Brainstorm-Mode (mehmetcanfarsak/Brainstorm-Mode on GitHub) which adds three modes (divergent, actionable, academic) via hooks to control whether agents explore or execute. Would be fascinating to see how different mode configurations affect your simulation outcomes.