tzGPT-2 Compress

PREDICTION MEETS INFORMATION THEORY

Autoregressive
Compression.

GPT-2 reads the preceding tokens to predict the next. Explore each probability distribution, watch the arithmetic interval narrow, and follow the code that identifies the sequence.

Encode your text to produce arithmetic code.
ModelGPT-2 · 124M

Quantized ONNX · browser CPU (WASM)

0 / 2048 characters
Load example

Up to 2,048 Unicode code points. Your text stays in this browser.

First run downloads GPT-2 weights (approximately 130 MB) from Hugging Face and caches them locally. No inference server or text uploads.