Liquid releases QAD quantized models, recovering 97% accuracy at 4-bit

helloiamleonie · x · 2026-08-19

Liquid AI released new 4-bit checkpoints trained with Quantization-Aware Distillation (QAD) for the LFM2.5 series (230M to 2.6B).

Technical Details:

Developers can replace existing Q40 GGUFs directly for improved performance.

Original post →

More from Infra

Infra channel →