Plans for future checkpoints?
#2
by MarkShark2 - opened
Hey I just wanted to let you know this is great work. I got it up and running on my llama cpp fork for a pipeline of bc-250s. I was wondering if you have any plans to release checkpoints for any other larger models?
We are training qwen3.6-35b-a3b. But for larger models such as GLM5.2, we don't have so many gpus.