I got it running using Colab Pro+ (immediately got a V100 40GB VRAM GPU) - the 7... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		arthurcolle on March 4, 2023 \| parent \| context \| favorite \| on: Facebook LLAMA is being openly distributed via tor... I got it running using Colab Pro+ (immediately got a V100 40GB VRAM GPU) - the 7B model works with batch size of 8 and a max seq len of 1024

koheripbal on March 4, 2023 [–]

Sure, but the real value here is the 65B. Can you have multiple GPUs on colab?

arthurcolle on March 4, 2023 | [–]

I can't even get the 13B on colab to do inference with a very small sequence length.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact