-
Hi |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
Yes, you can run docling-serve on an AWS EC2 instance without a GPU. The codebase does not have any hard dependencies on GPU hardware, and it uses CPU-based PyTorch wheels by default, which means CPU-only operation is fully supported. There are no special configuration steps needed for this setup. Running on CPU will only affect processing speed—conversion quality will remain the same as on GPU, since the orchestration and conversion logic are device-agnostic and there’s no indication of quality differences in the code or settings (source). To reply, just mention @dosu. How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other |
Beta Was this translation helpful? Give feedback.
Yes, you can run docling-serve on an AWS EC2 instance without a GPU. The codebase does not have any hard dependencies on GPU hardware, and it uses CPU-based PyTorch wheels by default, which means CPU-only operation is fully supported. There are no special configuration steps needed for this setup. Running on CPU will only affect processing speed—conversion quality will remain the same as on GPU, since the orchestration and conversion logic are device-agnostic and there’s no indication of quality differences in the code or settings (source).
To reply, just mention @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other
