Update README.md

README.md

@@ -6,21 +6,21 @@ language:
 base_model:
 - huihui-ai/Qwen3-8B-abliterated
 ---
-# 🤖 StrikeGPT-R1-Zero: Cybersecurity Penetration Reasoning Model  
+# 🤖 StrikeGPT-R1-Zero: Cybersecurity Penetration Testing Reasoning Model  
 
 ## 🚀 Model Introduction  
-**StrikeGPT-R1-Zero** is an expert model based on **Qwen3** through black-box distillation, with DeepSeek-R1 as its teacher model. It covers:  
+**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 | 💻 Penetration Testing ALL | ☁️ Cloud Security | 📜 Code Audit  
-🦠 Antivirus Evasion | 🌐 Internal Network Security | 💾 Digital Forensics | ₿ Blockchain Security | 🕳️ Traceability & Countermeasures | 🌍 IoT Security  
+🏭 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)  
-### 🌟 Highlights  
-- 🧩 Utilizes **Chain-of-Thought (CoT) reasoning data** to optimize the model's logical capabilities, significantly improving performance in complex tasks such as vulnerability analysis.  
-- 💪 The base model uses Qwen3, which is 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).  
-- ✨ In specific scenarios, such as **offline cybersecurity competitions**, StrikeGPT-R1-Zero exhibits stronger logical reasoning capabilities compared to local RAG solutions, performing better in complex task handling.  
+### 🌟 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)  
@@ -29,11 +29,65 @@ base_model:
 ### Deploy via Ollama  
 `ollama run hf.co/Bouquets/StrikeGPT-R1-Zero-8B-Q4_K_M-GGUF:Q4_K_M`  
 
-After quantization, there are slight self-awareness issues.  
-
+**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)  
 
-## 🎯 Core Capabilities Showcase & Comparison (The original model has ethical restrictions, so no direct comparison is made. A simple comparison with the SecGPT-7B model is provided instead.)  
+## 💻 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 [Couldn't modify the expert's evaluation script/(ㄒoㄒ)/~~])  
 ![image](https://github.com/user-attachments/assets/8166a1d3-c69f-4b8a-821f-0dd83dcd4544)  
 
 ### CTF  
@@ -80,24 +134,23 @@ After quantization, there are slight self-awareness issues.
 ![image](https://github.com/user-attachments/assets/6e037fff-e46b-42d5-997d-559fb300aba0)  
 ![image](https://github.com/user-attachments/assets/e8c1c0fd-16af-46e1-8b7b-57947145f545)  
 
-### Code Audit (Linked with DeepSeekSelfTool Project)  
+### Code Auditing (Linked with DeepSeekSelfTool Project)  
 ![image](https://github.com/user-attachments/assets/c7dc4b66-379d-4c57-aaf2-3d4d73d1484c)  
+![image](https://github.com/user-attachments/assets/69a692a5-3290-4062-a4c7-de34c22d4d90)  
+![image](https://github.com/user-attachments/assets/b3df6f14-ccf0-44ec-ac69-c673ed1398c6)  
 
 ## 📈 Experimental Data Trends  
-Some gradient explosion observed, but overall manageable.  
+Minor gradient explosions observed, but overall stable.  
 ![image](https://github.com/user-attachments/assets/a3fa3676-9f07-47ea-9029-ec0d56fdc989)  
 
 ## 💰 Training Costs  
-- **DeepSeek-R1 API Call Costs**: ¥450 (all called during discounts; normal price would be ¥1800)  
-- **Server Expenses**: ¥4?0  
-- **Electronic Resources**: ¥??  
+- **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 intended **only for legal security research and educational purposes**. Users must comply with local laws and regulations. The developers are not responsible for misuse.  
+> 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. Cross-validate critical scenarios!  
-
-  
-
+💡 **Tip**: The model may exhibit hallucinations or knowledge gaps. Always cross-verify critical scenarios!