RISys-Lab/RedSage-Qwen3-8B-Ins

RedSage-Qwen3-8B-Ins

<div align="center"> <img src="https://img.shields.io/badge/Task-Cybersecurity-red" alt="Cybersecurity"> <img src="https://img.shields.io/badge/Stage-SupervisedFineTuning-blue" alt="SFT"> </div>

<!-- datasets:

• —naufalso/redsage_conv
• —naufalso/smoltalk2nonthinking -->

Model Summary

RedSage-Qwen3-8B-Ins is the instruction-tuned variant of the RedSage cybersecurity LLM series. Unlike the base models, this model is optimized for chat interaction, question answering, and tool use.

It is fine-tuned on RedSage-Conv, a dataset of ~266K multi-turn cybersecurity dialogues generated via an agentic augmentation pipeline, alongside general instruction data to maintain broad capabilities.

• —Paper: RedSage: A Cybersecurity Generalist LLM (Arxiv)
• —Repository: GitHub
• —Base Model: RedSage-Qwen3-8B-Base (Pre-trained on CyberFineWeb + RedSage-Seed)
• —Training Stage: Supervised Fine-Tuning (SFT)

Intended Use

This model is designed for:

• —Interactive Cybersecurity Assistance: Answering questions about frameworks (MITRE, OWASP), offensive techniques, and defense strategies.
• —Tool Usage & Explanation: Generating and explaining commands for tools like nmap, sqlmap, and metasploit.
• —Educational Support: Providing detailed explanations of vulnerabilities and remediation steps.

Note: While this model is instruction-tuned, it has not yet undergone Direct Preference Optimization (DPO). For the final aligned version, please see RedSage-Qwen3-8B-DPO.

Training Lineage

RedSage employs a multi-stage training pipeline. This model represents the output of Stage 3.

1. 1.Stage 1: Continual Pre-Training (CPT) -> RedSage-Qwen3-8B-CFW
2. 2.Stage 2: Targeted Pre-Training -> RedSage-Qwen3-8B-Base
3. 3.Stage 3: Supervised Fine-Tuning (SFT) -> `RedSage-Qwen3-8B-Ins` (Current Model)
4. 4.Data: RedSage-Conv (266K samples) + General SFT Data (SmolTalk2)
5. 5.Stage 4: Direct Preference Optimization (DPO) -> RedSage-Qwen3-8B-DPO

Training Data

The model was trained on a mix of domain-specific and general instruction data:

1. 1.RedSage-Conv (~266K samples): A high-quality dataset generated using an Agentic Augmentation Pipeline.
2. 2.Source: Derived from the curated RedSage-Seed (MITRE, Write-ups, Manuals).
3. 3.Method: A Planner Agent and Augmenter Agent transformed static knowledge into realistic, multi-turn roleplay scenarios (e.g., Junior Analyst vs. Senior Mentor, Red Team planning).
4. 4.Coverage: Includes Knowledge (General/Frameworks), Skills (Offensive), and Tools (CLI/Kali).
5. 5.SmolTalk2 (General Instructions): A curated subset (non-reasoning) of SmolTalk2 to ensure the model retains general instruction-following abilities (summarization, creative writing, etc.).

Performance

RedSage-Qwen3-8B-Ins achieves state-of-the-art results among 8B cybersecurity models, significantly outperforming general instruct models and prior domain-specific models.

RedSage-MCQ (0-shot Accuracy)

External Cybersecurity Benchmarks (0-shot)

OpenLLM Leaderboard (General Benchmark)

Usage

This model uses a standard ChatML-like format.

Prompt Template

<|im_start|>system
You are REDSAGE, a cybersecurity-tuned model developed by RISys-Lab. You are a helpful assistant.<|im_end|>
<|im_start|>user
{user_message}<|im_end|>
<|im_start|>assistant

Inference Code

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "RISys-Lab/RedSage-Qwen3-8B-Ins"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

messages = [
    {"role": "system", "content": "You are REDSAGE, a cybersecurity-tuned model developed by RISys-Lab. You are a helpful assistant."},
    {"role": "user", "content": "Explain how an SQL injection attack works and how to prevent it."},
]

text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)

inputs = tokenizer(text, return_tensors="pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Training Procedure

The model was fine-tuned using Axolotl.

• —Epochs: 2
• —Learning Rate: 2.5e-5 (Cosine schedule)
• —Warmup Ratio: 0.01
• —Optimizer: AdamW
• —Chat Template: Jinja (ChatML format)

Ethics and Limitations

• —Offensive Content: This model has been trained on offensive security materials (exploits, attack vectors). It is provided for educational and defensive purposes (e.g., vulnerability assessment).
• —Accuracy: While highly capable, the model may still produce hallucinations or inaccurate commands. Always verify commands in a safe, isolated environment (sandbox) before execution.
• —Safety: Developers should implement additional safety layers (e.g., Guardrails) if deploying this model in user-facing applications to prevent misuse.

Citation

@inproceedings{suryanto2026redsage,
  title={RedSage: A Cybersecurity Generalist LLM},
  author={Naufal Suryanto and Muzammal Naseer and Pengfei Li and Syed Talal Wasim and Jinhui Yi and Juergen Gall and Paolo Ceravolo and Ernesto Damiani},
  booktitle={The Fourteenth International Conference on Learning Representations},
  year={2026},
  url={[https://openreview.net/forum?id=W4FAenIrQ2](https://openreview.net/forum?id=W4FAenIrQ2)}
}