---
title: mAIndlock
emoji: ๐ง
colorFrom: gray
colorTo: green
sdk: docker
app_port: 7860
pinned: true
license: mit
short_description: Escape room where every NPC is a mortal mind of tiny LLMs
models:
- openbmb/MiniCPM5-1B-GGUF
- openbmb/VoxCPM2
- nvidia/NVIDIA-Nemotron-3-Nano-4B-GGUF
tags:
- thousand-token-wood
- off-the-grid
- off-brand
- tiny-titan
- llama-champion
- llama-cpp
- openbmb
- minicpm
- nemotron
- agents
- game
- neuroscience
---
# ๐ง mAIndlock
**An escape room where the lock is a mind โ and you can kill it.**
โถ **[Watch the 100-second demo](https://youtu.be/zEwnVR1kTZU)** ยท ๐ฎ **[Play it live](https://huggingface.co/spaces/build-small-hackathon/maindlock)** ยท ๐ง **[Deliberation traces](https://huggingface.co/datasets/arbios/maindlock-brain-traces)**
Every character is not one chatbot with a personality prompt. It is a **hierarchy of six tiny
offline language models**, the way decision neuroscience says a brain is built. You don't crack
a code. You change a person's decision by reaching their fears and the memories they keep โ or
you burn their mind down trying.
Cruelty makes a mind ruminate, and rumination spends its finite **thinking tokens**. As they
burn, it **forgets** โ for good. At zero, the mind goes dark, taking everything it knew with it.
There is no reload. Only the next room, and your reputation travelling ahead of you.
> **Tokens are a lifespan here, not a context window.**
---
### โ๏ธ TL;DR for judges
- **Track โ Thousand Token Wood.** "A thousand tokens to think with" is literal: every mind
starts with **1000 thinking tokens of life**. Spend them cruelly and it dies.
- **OpenBMB ($10k).** **Six of the ~seven model calls per NPC turn are MiniCPM** โ the four
sensory brain regions (threat, memory, habit, cost), each a separate 1B call. MiniCPM *is* the
brain. (more under fear: the amygdala ruminates, firing extra calls that burn life) The
story's key-handover lines and the voiced demo are rendered with **OpenBMB VoxCPM2** TTS โ a
second model from the family, fully offline.
- **NVIDIA.** **Nemotron 3 Nano 4B** is the voice you actually argue with โ the dlPFC that turns
the integrated value into words.
- **Off the Grid / Llama Champion.** Pure **llama.cpp**, zero cloud APIs. Flip on airplane mode;
every mind keeps thinking. Token counts come from the runtime, so the life-burn is honest
accounting.
- **Off-Brand.** A custom canvas game served from FastAPI, with a Gradio block mounted at
[`/about`](/about) โ not a stock Gradio UI.
- **No CPU patience required:** menu โ **๐ Watch a mind** replays a *real* recorded session
instantly (zero model calls), so the six-region cascade, the token burn and a mind's death
are visible in ten seconds.
The brain is real, not a metaphor: [**docs/ARCHITECTURE.md**](https://github.com/Arbios/maindlock/blob/main/docs/ARCHITECTURE.md)
ยท read one mind's full deliberation in [**docs/TRACE.md**](https://github.com/Arbios/maindlock/blob/main/docs/TRACE.md).
---
## Every NPC is a brain, not a chatbot
Your words travel through six regions โ **amygdala โ hippocampus โ striatum โ ACC โ vmPFC โ
dlPFC** โ each one a real call to a small local model. The amygdala rates threat; the
hippocampus surfaces a memory and whether it leans trust or fear; the striatum weighs habit; the
ACC weighs cost. The **vmPFC integrates them deterministically into one value** โ so the number
you see in the skull *is* the number that moves the relationship; the panel can never lie about
the outcome. The dlPFC speaks it in character.
**Open the skull** mid-conversation (๐ง ) and watch the regions argue about you in real time โ
each showing its **conviction**, read straight from the model's own token entropy. *A hosted
chat API never exposes that. Only a local mind can.*
Grounding is modern decision neuroscience โ the **value-based network** and the **dual-system**
(model-free habit vs. model-based goal) accounts โ deliberately **not** the debunked triune
"lizard brain." Acute stress shifts control from goal to habit (Schwabe & Wolf, 2009): that is
literally why lowering a character's fear unlocks their reasoning, and why fear burns life for
nothing.
## A life measured in words
Be cruel and the amygdala loops, spending tokens that move nothing. Each quarter of life lost
**burns a memory away** โ the hippocampus genuinely loses it, and the *Forgotten* panel shows
what's gone. Push far enough and the mind dies, leaving a savable **epitaph**: what it knew,
what it never got to tell. Be kind, keep the alarm quiet, and the mind spends almost nothing โ
empathy literally spares it.
## Code rules, the model dreams
Death, the key, rapport and the burn are **deterministic game state**. The small models
*generate* โ they never get to fabricate the outcome. A 4B voice can decorate the story's spine
but cannot rewrite who lives. That is what makes the stakes real instead of vibes.
## ๐ช Story mode โ ten rooms of one man's memory
A man wakes in a ward with no name. Every room is a fragment of his own mind; every NPC holds a
piece of who he is โ **alive โ home โ sorry โ trains โ โฆ** โ until the last room shows him the
one task he should never have solved. No character tells you what happened; each gives the piece
they know, and the tragedy assembles itself in your head, which is exactly where it takes place.
Past the story, **Endless** mode generates minds forever, and a built-in **level editor** lets
you rewrite any mind and save it.
## โ๏ธ Small models, doing the carrying
| Role | Model | Size |
|---|---|---|
| Sensory regions โ threat, memory, habit, cost (4 calls/turn) | **MiniCPM (OpenBMB)** | 1B |
| Voice (dlPFC) โ the words said aloud | **Nemotron 3 Nano (NVIDIA)** | 4B |
| Spoken voice โ story lines & demo (TTS) | **VoxCPM2 (OpenBMB)** | โ |
| Integration (vmPFC) | deterministic value network | 0 |
Runtime **llama.cpp** (the Space) / **Ollama** (laptop). Total weights **โค 5.3B**, fully offline.
## ๐
Eligibility โ feature โ lane
| Feature | Lane / badge |
|---|---|
| 6 of ~7 calls per NPC turn are MiniCPM (the four sensory regions) | **OpenBMB $10k** |
| OpenBMB VoxCPM2 TTS voices the story spine + the demo | **OpenBMB $10k** (family breadth) |
| Nemotron 3 Nano 4B as the executive voice (dlPFC) | **NVIDIA** |
| 1000-token life, fear-burn, death by token exhaustion | **Thousand Token Wood** |
| Pure llama.cpp, airplane-mode proof, no cloud APIs | **Off the Grid ยท Llama Champion** |
| Custom canvas front, Gradio mounted at `/about` | **Off-Brand** |
| Four MiniCPM-1B sensory regions per mind | **Tiny Titan** |
| [Brain-region deliberation traces](https://huggingface.co/datasets/arbios/maindlock-brain-traces) published on the Hub | **Sharing is Caring** |
| [Department LoRA fine-tune](https://huggingface.co/arbios/mindlock-minicpmv46-departments-lora) of MiniCPM-V 4.6 (sensory-region behavior) | **Well-Tuned** |
| Two write-ups โ [the neuroscience](https://huggingface.co/blog/build-small-hackathon/brain-regions-science) ยท [the engineering](https://huggingface.co/blog/build-small-hackathon/minds-offline-engineering) | **Field Notes** |
| Intra-NPC six-region cognitive hierarchy | **Best Agent** |
## How to play
- **WASD / arrows** move ยท **E** talk / use the door ยท **๐ง or /brain** open the skull
- Listen, learn the word that reaches each keeper, and say it like you mean it.
- In a hurry? Menu โ **๐ Watch a mind** โ a real recorded session, no waiting.
---
๐ค *Built small for the **Hugging Face ร Gradio** hackathon โ Thousand Token Wood track.
Minds by **OpenBMB MiniCPM** and **NVIDIA Nemotron**, on **llama.cpp**, fully offline.
Original soundtrack by **DjinAscet**.*
**Links** โ โถ [Demo video](https://youtu.be/zEwnVR1kTZU) ยท ๐ป [Code](https://github.com/Arbios/maindlock) ยท ๐ [Traces dataset](https://huggingface.co/datasets/arbios/maindlock-brain-traces) ยท ๐ง [Dept LoRA](https://huggingface.co/arbios/mindlock-minicpmv46-departments-lora) ยท โ๏ธ Blog: [the science](https://huggingface.co/blog/build-small-hackathon/brain-regions-science) ยท [the engineering](https://huggingface.co/blog/build-small-hackathon/minds-offline-engineering) ยท ๐ฃ Social: [X](https://x.com/Arbios/status/2065982364635787774) ยท [LinkedIn](https://www.linkedin.com/posts/arbi_maindlock-a-hugging-face-space-by-build-small-hackathon-share-7471745737681465344-6YHz/)