---
title: "S&Box LLM Proof of Concept - s&box game"
description: "Proof of concept for running a real LLM entirely inside s&box using a custom C# GPT-Neo inference engine. No APIS, fully local"
canonical: "https://sbox.watch/fr/games/jeffskitchen/llm_poc/"
generated_at: "2026-10-09T17:38:31.388Z"
---

# S&Box LLM Proof of Concept

Proof of concept for running a real LLM entirely inside s&box using a custom C\# GPT-Neo inference engine. No APIS, fully local

- Type: jeu
- Identifiant: jeffskitchen.llm_poc
- Créateur: jeffskitchen
- URL canonique: https://sbox.watch/fr/games/jeffskitchen/llm_poc/
- Markdown: https://sbox.watch/fr/games/jeffskitchen/llm_poc/index.md
- sbox.game: https://sbox.game/jeffskitchen/llm_poc/
- Généré: 2026-10-09T17:38:31.388Z

## Statistiques

- Joueurs en ce moment: 0
- pic de joueurs sur 24 h: 0
- dynamique sur 24 h: +0
- Terry Score: 13%
- Favoris: 4
- Votes positifs: 3
- Votes négatifs: 7
- Nombre total de joueurs: 25
- Nombre total de sessions: 29
- Temps de jeu cumulé: 1,7 heures
- Temps de jeu gagné sur les 7 derniers jours: 3 minutes
- Valeur de référence sur 7 jours observée: 2026-10-02T06:49:21.208Z
- Mis à jour: 17 août 2026
- Créé: 17 août 2026
- Taille: 32.1 MB
- Fichiers: 41
- Tags: llm

## Description

!!!!NOTE!!!!!!:
This is not meant to be even a tiny bit close to an actual useable ChatGPT or Claude clone, the model (TinyStories-Instruct-1M) is quite literally 1/2000 the size of any flagship model. The findings in this project can be used to build larger language model projects.

S&Box LLM POC is a proof of concept demonstrating that a transformer language model can run entirely inside the s&box game runtime without external AI software, servers, APIs, Python, Ollama, or separate executables.

The project includes a custom C\# inference engine capable of loading packaged model weights, GPT-2 tokenization, GPT-Neo transformer inference, global and local attention, and autoregressive text generation directly within s&box.

The current implementation runs TinyStories-Instruct-1M locally on the CPU and has been numerically validated against the original PyTorch/Hugging Face model. Generated token sequences match the reference implementation while inference runs on an s&box worker thread to keep the game responsive.

The end goal of the POC is a simple in-game chat interface powered entirely by an LLM running inside s&box itself.

From this, I hope other developers can see what is possible and potentially use small LLM to power NPCs, game functions or even whole games.




## More by jeffskitchen

| Package | Identifiant | Joueurs | Favoris | Votes positifs | Score | Markdown |
| --- | --- | --- | --- | --- | --- | --- |
| AI Decision Maker (GLiNER 2.5) | jeffskitchen.aidecisionmakergliner25 | 0 | 1 | 0 | -- | https://sbox.watch/fr/games/jeffskitchen/aidecisionmakergliner25/index.md |