Apply for community grant: Personal project (gpu)

#1
by Spestly - opened

My Project (Atlas) is a model that is based on Deepseek R1 Qwen Distills. As of now, the model outperforms its based model in 6/7 Metric (Can be verified on the open LLM leaderboard). However, due to hardware constraints, I can only allow users to run 1.5B and below. I don't need anything High-end and fancy, but just enough to run a 7B model and maybe a 14B model.

newplot.png

Sign up or log in to comment