Bringing Nunchaku 4-bit Diffusion Inference to Diffusers [LAB]
This technical update introduces efficient 4-bit quantization for diffusion models, significantly optimizing inference performance. https://huggingface.co/blog/nunchaku-diffusers (huggingface.co)
New Models Today — AI & LLM Releases - Price Per Token [web]
Recent releases include MoonshotAI's Kimi K3 and Meta's Muse Spark 1.1, highlighting the rapid pace of model iteration. https://pricepertoken.com/news/model-releases (pricepertoken.com)