The full-sized Hy3 model is 598GB on Hugging Face, the FP8 quantized version is 300GB, and it supports a 256K context length.

Hy3 is a large download even quantized, but supports a long 256K-token context window. ✦ AI generated

Simon Willison · Simon Willison's Weblog · 2026-07-06 · original ↗

The full-sized model is 598GB on Hugging Face, and the FP8 quantized one is 300GB. The context length is 256K.

Read full article ↗excerpt · fair-use quotation