Show HN: Prompt injection detector beats ProtectAI by 19% accuracy, 8.9x smaller
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Summary not yet generated for this article.
Original Text
ProtectAI's deberta-v3 is the most downloaded prompt injection classifier on HuggingFace. We benchmarked directly against it on Qualifire's independent dataset (excluded from our training).
* 91.68% vs 72.28% accuracy
* 95.84% vs 65.33% precision — they block 1 in 3 legitimate users
* 83MB vs 739MB ONNX size
* 101ms vs 646ms on CPU
Comments (2 of ) ⌛