LinHekaka/RedSage-Qwen3-8B-Ins

TEXT GENERATIONPricing:Input $0.468 / Output $1.82Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 18, 2026Architecture:Transformer Featherless Exclusive Cold

LinHekaka/RedSage-Qwen3-8B-Ins is an 8 billion parameter instruction-tuned variant of the RedSage cybersecurity LLM series, developed by RISys-Lab. Optimized for chat interaction, question answering, and tool use, it is fine-tuned on a 266K multi-turn cybersecurity dialogue dataset. This model excels at providing interactive cybersecurity assistance, generating tool commands, and offering educational support for vulnerabilities and remediation.

Loading preview...

Model Overview

RedSage-Qwen3-8B-Ins is an 8 billion parameter instruction-tuned model from the RedSage cybersecurity LLM series, developed by RISys-Lab. It is specifically optimized for chat interaction, question answering, and tool use within the cybersecurity domain. This model is the result of Stage 3 (Supervised Fine-Tuning) in the RedSage multi-stage training pipeline, building upon the RedSage-Qwen3-8B-Base model.

Key Capabilities

  • Interactive Cybersecurity Assistance: Provides answers on frameworks like MITRE and OWASP, offensive techniques, and defense strategies.
  • Tool Usage & Explanation: Generates and explains commands for cybersecurity tools such as nmap, sqlmap, and metasploit.
  • Educational Support: Offers detailed explanations of vulnerabilities and steps for remediation.
  • Strong Performance: Achieves state-of-the-art results among 8B cybersecurity models, significantly outperforming general instruct models and prior domain-specific models on benchmarks like RedSage-MCQ and various external cybersecurity benchmarks.

Training Details

The model was fine-tuned on RedSage-Conv, a dataset of approximately 266,000 multi-turn cybersecurity dialogues generated via an agentic augmentation pipeline. This dataset covers knowledge, skills (offensive), and tools (CLI/Kali). It also incorporated a curated subset of general instruction data (SmolTalk2) to maintain broad instruction-following abilities.

Intended Use Cases

This model is ideal for developers and security professionals seeking an LLM specialized in:

  • Building interactive cybersecurity assistants.
  • Automating explanations of security concepts and tools.
  • Enhancing educational platforms with detailed vulnerability insights.

Note: While highly capable, users should verify generated commands in a safe, isolated environment and consider implementing additional safety layers for deployment.

Finetunes
1 models