FP8?
#3
by gearwave00001 - opened
Please consider creating an 8B FP8 version. I use a dedicated 5060ti 16GB to run Qwen3-embedding-8B-FP8 and would like to try the new nemotron, but I don't want to give up my larger GPUs for embedding tasks.
Hi @gearwave00001 ! Thanks for the interest in the model. However, we don't have immediate plans to release FP8 version of the 8B model.
ybabakhin changed discussion status to closed