Add SAAP Performance Benchmark Interface

README.md

@@ -1,12 +1,20 @@
 ---
-title: Ollama Test
-emoji: 🏃
-colorFrom: yellow
-colorTo: yellow
+title: SAAP Performance Benchmark
+emoji: 🚀
+colorFrom: blue
+colorTo: purple
 sdk: gradio
-sdk_version: 5.47.2
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# SAAP Ollama Performance Benchmark
+
+Master Thesis: satware AI Autonomous Agent Platform
+
+Test your local Ollama models with SAAP-specific scenarios:
+- Multi-Agent Coordination
+- On-Premise Performance Analysis  
+- Agent Role Simulations
+- Real-time Response Benchmarking

app.py

@@ -0,0 +1,183 @@
+import gradio as gr
+import requests
+import json
+import time
+from datetime import datetime
+
+class OllamaSAAPBenchmark:
+    def __init__(self, base_url="http://localhost:11434"):
+        self.base_url = base_url
+        
+    def test_agent_response(self, prompt, model, agent_role="General"):
+        """Test Agent-spezifische Responses für SAAP"""
+        
+        # SAAP-spezifische Prompts je nach Agent-Rolle
+        saap_prompts = {
+            "Jane": f"Als KI-Architektin: {prompt}",
+            "John": f"Als Entwickler: {prompt}", 
+            "Justus": f"Als Rechtsexperte: {prompt}",
+            "General": prompt
+        }
+        
+        final_prompt = saap_prompts.get(agent_role, prompt)
+        start_time = time.time()
+        
+        try:
+            response = requests.post(
+                f"{self.base_url}/api/generate",
+                json={
+                    "model": model,
+                    "prompt": final_prompt,
+                    "stream": False,
+                    "options": {"temperature": 0.7, "num_predict": 256}
+                },
+                timeout=60
+            )
+            
+            end_time = time.time()
+            
+            if response.status_code == 200:
+                result = response.json()
+                return {
+                    "response": result.get("response", ""),
+                    "time": f"{end_time - start_time:.2f}s",
+                    "model": model,
+                    "agent_role": agent_role,
+                    "tokens": len(result.get("response", "").split()),
+                    "status": "✅ Success"
+                }
+            else:
+                return {"status": f"❌ Error {response.status_code}", "time": f"{end_time - start_time:.2f}s"}
+                
+        except Exception as e:
+            return {"status": f"❌ Connection Error: {str(e)[:50]}...", "time": f"{time.time() - start_time:.2f}s"}
+
+    def list_models(self):
+        try:
+            response = requests.get(f"{self.base_url}/api/tags")
+            if response.status_code == 200:
+                models = response.json().get("models", [])
+                return [model["name"] for model in models]
+            return ["Connection failed - check if Ollama is running"]
+        except:
+            return ["❌ Cannot connect to Ollama"]
+
+# Initialize benchmark system
+benchmark = OllamaSAAPBenchmark()
+available_models = benchmark.list_models()
+
+# SAAP Benchmark Interface
+def run_saap_benchmark(prompt, selected_models, agent_role):
+    if not prompt.strip():
+        return "⚠️ Bitte geben Sie einen Test-Prompt ein."
+    
+    results = []
+    results.append(f"# 🚀 SAAP Multi-Agent Performance Benchmark")
+    results.append(f"**Agent Role:** {agent_role}")
+    results.append(f"**Test Prompt:** {prompt}")
+    results.append(f"**Models:** {', '.join(selected_models)}")
+    results.append(f"**Timestamp:** {datetime.now().strftime('%H:%M:%S')}")
+    results.append("---")
+    
+    total_time = 0
+    for model in selected_models:
+        if model in available_models:
+            result = benchmark.test_agent_response(prompt, model, agent_role)
+            
+            results.append(f"## 🤖 {model.upper()} ({agent_role})")
+            results.append(f"**Status:** {result.get('status', '❌ Error')}")
+            results.append(f"**Response Time:** {result.get('time', 'N/A')}")
+            results.append(f"**Tokens Generated:** {result.get('tokens', 0)}")
+            
+            if 'response' in result and result['response']:
+                preview = result['response'][:100].replace('\n', ' ')
+                results.append(f"**Response Preview:** {preview}...")
+            
+            results.append("---")
+            
+            # Add to total time for averages
+            try:
+                time_val = float(result.get('time', '0').rstrip('s'))
+                total_time += time_val
+            except:
+                pass
+    
+    # Performance Summary
+    if selected_models:
+        avg_time = total_time / len(selected_models)
+        results.append(f"## 📊 Performance Summary")
+        results.append(f"**Average Response Time:** {avg_time:.2f}s")
+        results.append(f"**Total Models Tested:** {len(selected_models)}")
+        
+        # SAAP Performance Assessment
+        if avg_time < 2.0:
+            results.append(f"**SAAP Assessment:** ✅ Excellent for real-time multi-agent coordination")
+        elif avg_time < 5.0:
+            results.append(f"**SAAP Assessment:** ⚠️ Acceptable for batch processing")
+        else:
+            results.append(f"**SAAP Assessment:** ❌ Too slow for interactive agents")
+    
+    return "\n".join(results)
+
+# Gradio Interface
+with gr.Blocks(title="SAAP Performance Benchmark", theme=gr.themes.Soft()) as demo:
+    gr.Markdown("# 🚀 SAAP - satware AI Agent Platform Benchmark")
+    gr.Markdown("**Master Thesis:** Hanan Wandji Danga | **Hochschule Worms** | **satware AG**")
+    
+    with gr.Row():
+        with gr.Column(scale=2):
+            prompt_input = gr.Textbox(
+                label="SAAP Test Prompt",
+                placeholder="Beispiel: Entwickle eine Systemarchitektur für Multi-Agent Koordination",
+                lines=3,
+                value="Erkläre die Vorteile einer On-Premise Multi-Agent-Plattform gegenüber Cloud-Lösungen."
+            )
+            
+            agent_role = gr.Dropdown(
+                choices=["General", "Jane", "John", "Justus"],
+                label="Agent Role Simulation",
+                value="General"
+            )
+            
+        with gr.Column(scale=1):
+            model_selection = gr.CheckboxGroup(
+                choices=available_models,
+                label="Models to Benchmark",
+                value=available_models[:2] if len(available_models) >= 2 else available_models
+            )
+            
+            benchmark_btn = gr.Button("🚀 Run SAAP Benchmark", variant="primary", size="lg")
+    
+    # Results
+    results_output = gr.Markdown(label="Benchmark Results")
+    
+    # Benchmark function
+    benchmark_btn.click(
+        run_saap_benchmark,
+        inputs=[prompt_input, model_selection, agent_role],
+        outputs=results_output
+    )
+    
+    # System Info
+    with gr.Accordion("ℹ️ System Information", open=False):
+        gr.Markdown(f"""
+        ### 📋 SAAP Test Environment
+        - **Available Models:** {len(available_models)}
+        - **Models:** {', '.join(available_models)}
+        - **Ollama Server:** {benchmark.base_url}
+        
+        ### 🎯 SAAP Performance Targets
+        - **Real-time Coordination:** < 2s per response
+        - **Batch Processing:** < 5s per response
+        - **Multi-Agent Sync:** < 10s for complex workflows
+        
+        ### 🎓 Master Thesis Context
+        **Projekt:** SAAP - satware AI Autonomous Agent Platform  
+        **Student:** Hanan Wandji Danga  
+        **Universität:** Hochschule Worms  
+        **Betreuung:** Michael Wegener  
+        **Ziel:** On-Premise Multi-Agent-Plattform mit lokalen LLMs
+        """)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,2 @@
+gradio>=4.0.0
+requests>=2.31.0