Llama-3.1-FoundationAI-SecurityLLM-Reasoning-8B Technical Report

2026-01-28

AI Summary

This work addresses the current lack of open-source, natively reasoning-capable large language models tailored for cybersecurity. Building upon Llama-3.1-8B-Base, we propose and release the first open-source large language model with native support for cybersecurity reasoning. The model is trained via a two-stage process—supervised fine-tuning (SFT) followed by reinforcement learning with verifiable rewards (RLVR)—integrating datasets spanning cybersecurity analysis, instruction following, and mathematical reasoning. It further incorporates system prompts and safety guardrails to enhance reliability. Evaluated across 10 cybersecurity-specific and 10 general-purpose benchmarks, the model demonstrates superior performance in multi-hop reasoning and security-related tasks, significantly outperforming comparable models and rivaling much larger counterparts while maintaining a strong balance between domain expertise and general capabilities.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Large Vision Models
• Natural Language Processing: (Large) Language Models

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

We present Foundation-Sec-8B-Reasoning, the first open-source native reasoning model for cybersecurity. Built upon our previously released Foundation-Sec-8B base model (derived from Llama-3.1-8B-Base), the model is trained through a two-stage process combining supervised fine-tuning (SFT) and reinforcement learning from verifiable rewards (RLVR). Our training leverages proprietary reasoning data spanning cybersecurity analysis, instruction-following, and mathematical reasoning. Evaluation across 10 cybersecurity benchmarks and 10 general-purpose benchmarks demonstrates performance competitive with significantly larger models on cybersecurity tasks while maintaining strong general capabilities. The model shows effective generalization on multi-hop reasoning tasks and strong safety performance when deployed with appropriate system prompts and guardrails. This work demonstrates that domain-specialized reasoning models can achieve strong performance on specialized tasks while maintaining broad general capabilities. We release the model publicly at https://huggingface.co/fdtn-ai/Foundation-Sec-8B-Reasoning.

Problem

• cybersecurity
• reasoning
• foundation model
• domain specialization
• generalization

Innovation

• reasoning model
• cybersecurity
• reinforcement learning from verifiable rewards
• supervised fine-tuning
• domain-specialized LLM