tashrifrahman/TR-DeepSeek-R1-Phishing-Detector
TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Aug 7, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold
The TR-DeepSeek-R1-Phishing-Detector is an 8 billion parameter Llama-based model developed by tashrifrahman, fine-tuned from unsloth/deepseek-r1-distill-llama-8b-unsloth-bnb-4bit. This model is specifically optimized for phishing detection tasks, leveraging efficient training with Unsloth and Huggingface's TRL library. It is designed to provide specialized performance in identifying phishing attempts, making it suitable for security-focused applications.
Loading preview...
TR-DeepSeek-R1-Phishing-Detector Overview
This model, developed by tashrifrahman, is an 8 billion parameter Llama-based language model specifically fine-tuned for phishing detection. It originates from the unsloth/deepseek-r1-distill-llama-8b-unsloth-bnb-4bit base model.
Key Capabilities
- Specialized Phishing Detection: The model is explicitly trained to identify and classify phishing attempts, making it a focused tool for cybersecurity applications.
- Efficient Training: It was fine-tuned using Unsloth and Huggingface's TRL library, which enabled a 2x faster training process.
- Llama Architecture: Built upon the Llama architecture, it benefits from the robust capabilities and widespread support of this model family.
Good For
- Security Applications: Ideal for integrating into systems requiring automated detection of phishing emails, messages, or URLs.
- Research in Cybersecurity: Useful for researchers exploring advanced methods for identifying malicious content.
- Developers Seeking Optimized Models: Provides a pre-trained, specialized model for a critical security task, potentially reducing development time for related features.