updated work flow , just install missing node and extract this inside clip folder
https://civitai.com/models/652981/gguf-workflow-simple?modelVersionId=748334
🎨 Introducing the T5 Model in GGUF Format! 🎨
Are you tired of relying on your GPU for high-performance AI inference? Look no further than the latest T5 model in the efficient GUFF format! 💻
After careful consideration, the best options are:
🔵 t5-v1_1-xxl-encoder-Q8_0.gguf 🔵
this the most closed to fp16 what ever your GPU
GGUF offers exceptional quantization capabilities, allowing for faster and more memory-efficient inference on both CPUs and GPUs. 🚀 That means top-notch performance without heavy GPU requirements. 💪
The other options, t5-v1_1-xxl-encoder-Q4_K_M.gguf and t5-v1_1-xxl-encoder-Q5_K_M.gguf, are also great, but the Q8 version truly shines in terms of efficiency and versatility. 🌟
was start here
thanks for city96 here the full list of all t5 guff models and dev&schnell guff models