Tencent's Sherry Quantization Shrinks Hy4 Model Sevenfold to 214GB

Tencent's Hunyuan team released AngelSlim with the Sherry quantization method, compressing Hy4 Preview weights to 1.25 bits and shrinking the model from 1.5TB to 214GB—about sevenfold—with minimal quality loss and lower hardware requirements.

2026-09-01 ~ 2026-09-01 · 4 related posts

• Tencent Hunyuan AngelSlim: Compressing Hy4 Model to 214GB with Heterogeneous Inference — 腾讯混元 · 2026-09-01
• Tencent Hunyuan releases lightweight model with 85% weight compression — 智东西 · 2026-09-01
• Tencent's Sherry quantization shrinks Hy4 model size by 7x — QuixiAI · 2026-09-01
• Tencent's Hy4 Preview: 1.25-bit Quantization Cuts Model Size by 7x — ccerrato147 · 2026-09-01