AI Model Performance Dashboard

Interactive visualization of llama3.3, gpt-oss:120b, and qwen3:235b performance metrics

1 16 32

Average Latency

2.34s

Best: llama3.3

Throughput

1,248/s

Best: qwen3:235b

Time to First Token

0.87s

Best: gpt-oss:120b

Latency vs Throughput

Time to First Token

Performance Heatmap

Multi-dimensional Analysis

Performance Comparison

Concurrency Impact

This dashboard features: 1. **Modern, responsive design** with gradient backgrounds and hover animations 2. **Interactive controls** for model selection, concurrency levels, and token sizes 3. **Multiple coordinated visualizations**: - Scatter plot showing latency vs throughput - Time-series analysis of time to first token - Performance heatmaps and parallel coordinates - Comparative bar charts and degradation curves 4. **Real-time filtering** and dynamic updates 5. **Performance metrics overview** with key insights 6. **Smooth animations** and professional styling 7. **Mobile-responsive layout** that works on tablets and desktops The dashboard uses Plotly.js for interactive charts and D3.js for custom visualizations, with a cohesive color scheme and professional UI components. All visualizations are coordinated and update based on user interactions.