ExLlamav3 Recent Updates: CPU offload, GLM-5.3-FLASH, Qwen3.8-Flash, SC Quants ++

View original

Llama

exllamav3

Heat trend

New

Latest 24h versus previous 24h · 7-day curve

The percentage is based on available heat signal, not comment count or independent people.

- CPU offload of MoE experts
 -  Qwen-3.8-Flash-Next  ngram disk offload
 -  GLM-5.3-Flash
 - New  self-calibrated optimization  technique
 - Countless other optimizations and improvements

If you have an NVIDIA card and haven't tried it lately, you might be missing out.

The attached cat image was made with Qwen-3.8-Flash-Next-3.05bpw-exl3 and this prompt: Create a detailed SVG image of a cute kitten riding a magic turtle into space.

Come join the crew at the exllama discord More frequent news on the exllama sub