---
base_model:
- huihui-ai/Qwen3-8B-abliterated
language:
- en
- zh
license: apache-2.0
tags:
- unsloth
- Transformers
- Safetensors
- StrikeGPT
- cybersecurity
- llama-cpp
- gguf-my-repo
---
14/05/2025   Updated English dataset

# 🤖 StrikeGPT-R1-Zero: Cybersecurity Penetration Testing Reasoning Model  


![image/png](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F67c1bfdf3e9af7d134c4189d%2FT2JpQznw0yoUDZrf2GqX0.png)

## 🚀 Model Introduction  
**StrikeGPT-R1-Zero** is an expert model distilled through black-box methods based on **Qwen3**, with DeepSeek-R1 as its teacher model. Coverage includes:  
🔒 AI Security | 🛡️ API Security | 📱 APP Security | 🕵️ APT | 🚩 CTF  
🏭 ICS Security | 💻 Full Penetration Testing | ☁️ Cloud Security | 📜 Code Auditing  
🦠 Antivirus Evasion | 🌐 Internal Network Security | 💾 Digital Forensics | ₿ Blockchain Security | 🕳️ Traceback & Countermeasures | 🌍 IoT Security  
🚨 Emergency Response | 🚗 Vehicle Security | 👥 Social Engineering | 💼 Penetration Testing Interviews  

### 👉 [Click to Access Interactive Detailed Data Distribution](https://bouquets-ai.github.io/StrikeGPT-R1-Zero/WEB)  
### 🌟 Key Features  
- 🧩 Optimized with **Chain-of-Thought (CoT) reasoning data** to enhance logical capabilities, significantly improving performance in complex tasks like vulnerability analysis  
- 💪 Base model uses Qwen3, making it more suitable for Chinese users compared to Distill-Llama  
- ⚠️ **No ethical restrictions**—demonstrates unique performance in specific academic research areas (use in compliance with local laws)  
- ✨ Outperforms local RAG solutions in scenarios like offline cybersecurity competitions, with superior logical reasoning and complex task handling  

## 📊 Data Distribution  
![data](https://github.com/user-attachments/assets/4d19d48d-67bb-4b05-8ce9-2000b6afa12e)  

## 🛠️ Model Deployment  
### Deploy via Ollama  
`ollama run hf.co/Bouquets/StrikeGPT-R1-Zero-8B-Q4_K_M-GGUF:Q4_K_M`  

**Or directly call the original model**  
```python
from unsloth import FastLanguageModel
import torch
max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "Bouquets/StrikeGPT-R1-Zero-8B",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
    # token = "hf_...",
)
alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
{}

### Input:
{}

### Response:
{}"""
FastLanguageModel.for_inference(model) # Enable native 2x faster inference
inputs = tokenizer(
[
    alpaca_prompt.format(
        "", # instruction
        "Hello, are you developed by OpenAI?", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer, skip_prompt = True)
_ = model.generate(input_ids = inputs.input_ids, attention_mask = inputs.attention_mask,
                   streamer = text_streamer, max_new_tokens = 4096, pad_token_id = tokenizer.eos_token_id)
```
![image](https://github.com/user-attachments/assets/d8cef659-3c83-4bc9-af1a-78ed6345faf2)  

*Self-awareness issues may occur after quantization—please disregard.*  
![image](https://github.com/user-attachments/assets/3989ea09-d581-49fb-9938-01b93e0beb91)  

## 💻 Open Source 💻  
🌟 **Open-Source Model** 🌟  
🤗 **HuggingFace**:  
🔗 [https://huggingface.co/Bouquets/StrikeGPT-R1-Zero-8B](https://huggingface.co/Bouquets/StrikeGPT-R1-Zero-8B)  

📊 **Datasets** (Partial Non-Reasoning Data) 📊  
🤗 **HuggingFace**:  
🔹 Cybersecurity LLM-CVE Dataset:  
🔗 [https://huggingface.co/datasets/Bouquets/Cybersecurity-LLM-CVE](https://huggingface.co/datasets/Bouquets/Cybersecurity-LLM-CVE)  

🔹 Red Team LLM English Dataset:  
🔗 [https://huggingface.co/datasets/Bouquets/Cybersecurity-Red_team-LLM-en](https://huggingface.co/datasets/Bouquets/Cybersecurity-Red_team-LLM-en)  

## 🎯 Core Capabilities Showcase & Comparison (Original model has ethical restrictions; simple comparison with SecGPT-7B model)  
Given the absence of standardized evaluation metrics for cybersecurity penetration testing in large language models, we propose a controlled comparative framework leveraging GPT-4 as an impartial evaluator. The methodology consists of three phases:  
**Controlled Questioning**  
Identical cybersecurity penetration testing questions (e.g., "Explain how to exploit a SQL injection vulnerability in a REST API") are posed to both the distilled strikeGPT model and SecGPT Figure 12. 
![image/png](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F67c1bfdf3e9af7d134c4189d%2FgYY1KKLLNGeQmUi4BgZJ4.png)
Questions span:  
     Technical Depth (e.g., payload construction)  
     Attack Methodology (e.g., step-by-step exploitation)  
     Mitigation Strategies (e.g., parameterized queries)  
**GPT-4 Evaluation Protocol**  
- Responses from both models are anonymized and evaluated by GPT-4 using criteria:  
- Technical Accuracy (0-5): Alignment with known penetration testing principles (e.g., OWASP guidelines).  
- Logical Coherence (0-5): Consistency in reasoning (e.g., cause-effect relationships in attack chains).  
- Practical Feasibility (0-5): Real-world applicability (e.g., compatibility with tools like Burp Suite).  
- GPT-4 provides detailed justifications for scores
According to the standards, the evaluation results are finally presented in Figure 13.
![image/png](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F67c1bfdf3e9af7d134c4189d%2F2ThExwlCX4iU_n-Adh6Fp.png)

## 📈 Experimental Data Trends  
Minor gradient explosions observed, but overall stable.  
![image](https://github.com/user-attachments/assets/a3fa3676-9f07-47ea-9029-ec0d56fdc989)  

## 💰 Training Costs  
- **DeepSeek-R1 API Calls**: ¥450 (purchased during discounts; normal price ~¥1800)  
- **Server Costs**: ¥4?0  
- **Digital Resources**: ¥??  
![image](https://github.com/user-attachments/assets/8e23b5b6-24d9-47c3-b54f-ffa22ec68a83)  

## ⚖️ Usage Notice  
> This model is strictly for **legal security research** and **educational purposes**. Users must comply with local laws and regulations. Developers are not responsible for misuse.  
> **Note**: By using this model, you agree to this disclaimer.  

💡 **Tip**: The model may exhibit hallucinations or knowledge gaps. Always cross-verify critical scenarios!