Is this text safe?
- 99.2%
- UCI SMS Spam Collection
- 104
- languages (mBERT)
- <1 s
- on GPU, SMPP-native
A family of SMS classifiers that run on your own servers. The Threat Classifier asks whether a text is safe. The new Category Classifier asks what kind of text it is.
The original OTS model is now the Threat Classifier. The Category Classifier, released October 2026, sits next to it.
Is this text safe?
What category of text is it?
Both run on your own servers: no message leaves your network. OTS-CC priority languages are English, Spanish, Indonesian, French, Hindi (including Hinglish), Arabic, Turkish, German and Dutch. Both APIs are documented in the TCXC AI API reference.
Independently benchmarked on recognized public datasets with a fully reproducible, offline evaluation harness. Our latest model update sharply improves modern smishing detection on the independent IMC 2025 corpus — closing the non-English gap while preserving best-in-class accuracy on classic spam.
Choose between Community Edition for development and testing, or Professional Edition for production-ready carrier-grade deployments.
Perfect for development, testing, and organizations with in-house AI expertise. Build and train your own models with complete transparency and control.
Enterprise-ready with pre-integrated SMPP stack, professionally trained models, and carrier-grade support. Deploy instantly into production environments.
OpenTextShield is completely open source, transparent, and community-driven
Full source code transparency with MIT license. No hidden algorithms, no vendor lock-in, complete control over your security infrastructure.
Built by telecom engineers for telecom operators. Contribute features, report issues, and shape the future of SMS security with the community.
Production-tested by major telecom operators. Scalable architecture, comprehensive documentation, and professional support available.

Multilingual, on-premise SMS classification — every message labelled ham, spam, or phishing in under 50 ms. The programmable SMSC turns those labels into operator policy.
"This is an A.I. use we can proudly get behind — flexible SMSC routing and filter rules + a dynamic open source A.I.-powered SMS firewall."
Both editions deliver sub-second spam detection with 99.2% benchmark accuracy on the UCI SMS Spam Collection, differing in deployment complexity and support level.
Production-ready Docker image with GPU acceleration and dynamic batching. Validated on AWS g4dn.4xlarge (NVIDIA T4) at ~100 messages/sec sustained — 132,000+ messages, zero failures, zero timeouts.
sudo apt update && sudo apt install -y nvidia-driver-535-server
sudo reboot
# After reboot — install NVIDIA Container Toolkit
sudo apt install -y nvidia-container-toolkit
sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker
# Sanity check — should list your GPU
docker run --rm --gpus all nvidia/cuda:12.6.0-base-ubuntu22.04 nvidia-smidocker pull telecomsxchange/opentextshield:v2.9.0
docker run -d \
--name ots \
--gpus all \
--restart unless-stopped \
-p 8002:8002 \
-p 8080:8080 \
telecomsxchange/opentextshield:v2.9.0# Confirm GPU acceleration
curl -s http://<host>:8002/metrics | grep ots_api_info
# Expected: ots_api_info{version="2.9.0",device="cuda",fp16="true",...} 1
# Watch live GPU utilization while sending traffic
nvidia-smi -l 2curl -X POST "https://ots.telecomsxchange.com/predict/" \
-H "accept: application/json" \
-H "Content-Type: application/json" \
-d '{"text":"URGENT! You have won $5000! Reply CLAIM to collect your prize money now! Limited time offer!","model":"ots-mbert"}'
# Response: {"label":"spam","probability":0.97,"processing_time":0.04,...}# Smaller GPUs (under 8 GB VRAM)
docker run -d --gpus all \
-e OTS_MAX_BATCH_SIZE=32 \
-e OTS_BATCH_WAIT_MS=50 \
-p 8002:8002 -p 8080:8080 \
telecomsxchange/opentextshield:v2.9.0
# Single-request debugging — disable batching
docker run --gpus all \
-e OTS_BATCHING_ENABLED=false \
-p 8002:8002 \
telecomsxchange/opentextshield:v2.9.0See why telecom operators choose OpenTextShield over expensive commercial AI APIs
Get started with OpenTextShield — explore the source, deploy locally, or talk to our team about the Professional Edition.