GENERATIVE LANGUAGE TRANSFORMER WITH CONTEXTUAL HIERARCHY

Build Your Own
AI Language Model

Train AI models from scratch. Join the distributed training network. Contribute GPU power to build something amazing together.

⚡ Get Started 🌐 Join the Hive
$ curl -sSL https://gltch.app/join | bash
╔═══════════════════════════════════════╗
║ GLTCH HIVE — Quick Join ║
╚═══════════════════════════════════════╝
📥 Cloning GLTCH...
🎮 GPU: NVIDIA RTX 4090
🚀 Starting training on GLTCH-10M...
4
Model Sizes
50M
Max Parameters
∞
Peers Welcome
MIT
Open Source

Features

🧠

Multiple Sizes

Train 2.7M, 10M, 25M, or 50M parameter models. Pick what fits your GPU.

📊

Live Dashboard

Watch training in real-time. See loss curves, speed, and generated samples.

💬

Chat Interface

Talk to your trained model with a web UI. Includes text-to-speech.

🌐

Distributed Training

Join the Hive network. Train across multiple machines and GPUs.

⚡

GPU Accelerated

Native CUDA support for NVIDIA GPUs. Also works on Apple Silicon.

📖

Learn AI

Understand transformers by building one from scratch. Full source code.

Model Sizes

GLTCH-2.7M
2.7M
~5 min on GPU
Good for learning
GLTCH-10M
10M
~15 min on GPU
Recommended
GLTCH-25M
25M
~30 min on GPU
Better quality
GLTCH-50M
50M
~1 hour on GPU
Best quality

Join the Hive

Contribute your GPU power to distributed training

🐝 Learn How to Join