Spam Detector

Runs inside your own AWS account: your text is processed on a SageMaker endpoint you control, inside your VPC, and is never sent to a third-party API.

This SageMaker model package provides a REST API to classify messages as spam or ham (legitimate).

The model is fine-tuned on SMS and email spam corpora and returns a label with a confidence score. Use it to filter user-generated content, protect messaging platforms, or pre-screen inbound email.

The API accepts input as JSON or CSV and supports real-time endpoints and batch transform.

We welcome your feedback at [email protected]

Model and training data The model is a ModernBERT-base (answerdotai/ModernBERT-base) text classifier fine-tuned with PyTorch and Hugging Face Transformers on the public sms_spam and SetFit/enron_spam corpora. It runs in network isolation on SageMaker, so no data leaves your account.

Known limitations - English only. Other languages are not supported. - The model was fine-tuned on inputs up to 512 tokens. Longer text is accepted, but accuracy is only characterized to that length; split long documents into passages before sending. - Trained on SMS and email text. Spam styles drift over time, so evaluate against your own recent traffic before relying on it.

Measured performance - Accuracy 99.1 percent, macro-F1 0.991, measured on 7,847 held-out examples from the SMS and Enron spam corpora. The split is seeded (70/20/10, seed 42) and was not used in training. - Real-time endpoint on ml.m5.large: 255 ms median per request (267 ms p95). - Throughput on the same instance: about 24 texts per second at a batch size of 32.