poolside
Laguna-M.1
text generationapache-2.0824.2K downloadsHugging Face
No server in stock right now has the 505 GB this model needs for inference.
GPU memory it needs (estimate)
Rule of thumb from the model's size. Real needs depend on batch size, context length and framework.
Inference (full precision)
505 GB
Inference (8-bit)
252 GB
Inference (4-bit)
126 GB
LoRA fine-tuning
547 GB
QLoRA (8-bit base)
273 GB
Full fine-tuning
2019 GB