Remove gpt-4

README.md

@@ -19,7 +19,7 @@ A simplified forensic image analysis system that uses an LLM agent to analyze im
 This project focuses on **agent flows and tool usage** rather than model-based classification. The agent:
 
 1. Receives an image directly
-2. Uses vision-capable LLM (e.g., GPT-4o) to analyze it
+2. Uses vision-capable LLM (e.g., GPT-5.1) to analyze it
 3. Can invoke forensic tools to gather additional evidence
 4. Provides reasoning and conclusions based on evidence
 
@@ -75,7 +75,7 @@ python scripts/analyze_image.py --image path/to/image.jpg --query "Is this image
 
 Use a different model:
 ```powershell
-python scripts/analyze_image.py --image path/to/image.jpg --model gpt-4o --temperature 0.3
+python scripts/analyze_image.py --image path/to/image.jpg --model gpt-5.1 --temperature 0.3
 ```
 
 ## Architecture

README_SPACE.md

@@ -4,7 +4,7 @@ AI-powered forensic image analysis system that detects AI-generated, manipulated
 
 ## 🚀 Features
 
-- **Vision Analysis**: Direct image analysis using GPT-4 Vision
+- **Vision Analysis**: Direct image analysis using GPT-5
 - **Forensic Tools**: Multiple forensic analysis tools including:
   - JPEG compression analysis
   - Frequency domain analysis (DCT/FFT)
@@ -59,7 +59,7 @@ OPENAI_API_KEY=your-key-here
 2. **Optional Query**: Add a specific question about the image (e.g., "Is this AI-generated?")
 3. **Configure Settings**: 
    - Enable/disable forensic tools
-   - Choose LLM model (gpt-4o recommended)
+   - Choose LLM model (gpt-5.1 recommended)
    - Adjust temperature
 4. **Analyze**: Click "🔍 Analyze Image" button
 5. **View Results**: See detailed analysis with visual description, forensic findings, and conclusion
@@ -67,10 +67,10 @@ OPENAI_API_KEY=your-key-here
 ## ⚙️ Configuration
 
 ### Models Supported
-- `gpt-4o` (recommended) - Latest vision model
-- `gpt-4o-mini` - Faster, more cost-effective
-- `gpt-4-turbo` - Previous generation
-- `gpt-4` - Standard GPT-4
+- `gpt-5.1` (recommended) - Latest vision model
+- `gpt-5` - Standard GPT-5
+- `gpt-5-mini` - Faster, more cost-effective
+- `gpt-5-nano` - Lightweight option
 
 ### Forensic Tools
 When enabled, the agent can use:
@@ -114,7 +114,7 @@ When enabled, the agent can use:
 - Check that all dependencies are installed correctly
 
 ### Memory Issues
-- Use `gpt-4o-mini` for lower memory usage
+- Use `gpt-5-mini` for lower memory usage
 - Disable forensic tools if running on CPU-only hardware
 
 ## 📝 License
@@ -124,6 +124,6 @@ When enabled, the agent can use:
 ## 🙏 Acknowledgments
 
 - Built with [LangChain](https://www.langchain.com/) and [LangGraph](https://github.com/langchain-ai/langgraph)
-- Powered by [OpenAI GPT-4 Vision](https://openai.com/)
+- Powered by [OpenAI GPT-5](https://openai.com/)
 - UI built with [Gradio](https://www.gradio.app/)
 

app.py

@@ -59,7 +59,7 @@ agent_config: dict = {}  # Store current agent configuration
 
 
 def initialize_agent(
-    model_name: str = "gpt-4o",
+    model_name: str = "gpt-5.1",
     temperature: float = 0.2,
     reasoning_effort: Optional[str] = None,
     use_tools: bool = True
@@ -105,7 +105,7 @@ def analyze_image(
     image,
     query: Optional[str] = None,
     use_tools: bool = True,
-    model_name: str = "gpt-4o",
+    model_name: str = "gpt-5.1",
     temperature: float = 0.2,
     reasoning_effort: Optional[str] = None,
 ):
@@ -301,7 +301,7 @@ def respond_chat(
     message: str,
     history: list,
     use_tools: bool = True,
-    model_name: str = "gpt-4o",
+    model_name: str = "gpt-5.1",
     temperature: float = 0.2,
     reasoning_effort: Optional[str] = None,
 ):
@@ -513,8 +513,8 @@ def create_interface():
                     
                     model_dropdown = gr.Dropdown(
                         label="LLM Model",
-                        choices=["gpt-4o", "gpt-4o-mini", "gpt-4-turbo", "gpt-4"],
-                        value="gpt-4o",
+                        choices=["gpt-5", "gpt-5.1", "gpt-5-mini", "gpt-5-nano"],
+                        value="gpt-5.1",
                         info="OpenAI model to use for analysis"
                     )
                     
@@ -574,7 +574,7 @@ def create_interface():
         <div style="text-align: center; padding: 20px; color: #666;">
             <p>
                 <strong>Forensic Image Analysis Agent</strong> | 
-                Powered by LangChain, LangGraph, and OpenAI GPT-4 Vision
+                Powered by LangChain, LangGraph, and OpenAI GPT-5
             </p>
             <p style="font-size: 0.9em;">
                 This agent analyzes images using vision-capable LLMs and forensic tools to detect
@@ -609,8 +609,8 @@ def create_chat_interface():
             ),
             gr.Dropdown(
                 label="LLM Model",
-                choices=["gpt-4o", "gpt-4o-mini", "gpt-4-turbo", "gpt-4"],
-                value="gpt-4o",
+                choices=["gpt-5", "gpt-5.1", "gpt-5-mini", "gpt-5-nano"],
+                value="gpt-5.1",
                 info="OpenAI model to use"
             ),
             gr.Slider(

scripts/analyze_image.py

@@ -31,8 +31,8 @@ def main():
     parser.add_argument(
         "--model",
         type=str,
-        default="gpt-4o",
-        help="LLM model to use (default: gpt-4o)"
+        default="gpt-5.1",
+        help="LLM model to use (default: gpt-5.1)"
     )
     parser.add_argument(
         "--temperature",

src/agents/forensic_agent.py

@@ -45,7 +45,7 @@ class ForensicAgent:
                  max_iterations: Optional[int] = 50):
         """
         Args:
-            llm_model: OpenAI model name (should support vision, e.g., gpt-4o)
+            llm_model: OpenAI model name (should support vision, e.g., gpt-5.1)
             temperature: LLM temperature
             reasoning_effort: Reasoning effort level for the model
             api_key: OpenAI API key (or set OPENAI_API_KEY env var)