LLM efficiency News and Analysis

2 verified stories covering LLM efficiency, product updates and industry developments.

On-device AI 2026-08-19

Liquid AI revives Q4_0 quantization with 97% accuracy via distillation

Liquid AI's QAD training method restores Q4_0 GGUF checkpoints to 96-97% of BF16 accuracy, preserving the format's speed edge on Arm CPUs and older devices.

#Open Source #AI inference optimization #LLM efficiency

Google 2026-04-20

Google TurboQuant compresses AI inference memory by 6x

Google Research's TurboQuant compresses KV Cache memory for large language models by at least 6x without retraining or accuracy loss, achieving up to 8x speedup on H100 GPUs.

#AI inference optimization #LLM efficiency #Technical breakthroughs