OBLITERATUS
DeepSeek-V4.1-Flash-1017-obliterated
text generationmit7K downloadsHugging Face
No server in stock right now has the 1687 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)
1687 GB
Inference (8-bit)
843 GB
Inference (4-bit)
422 GB
LoRA fine-tuning
1827 GB
QLoRA (8-bit base)
914 GB
Full fine-tuning
6747 GB