AI Language Model Benchmark Comparison

import matplotlib.pyplot as plt
import numpy as np

# AI model benchmarks
categories = ['Reasoning', 'Coding', 'Mathematics', 'Creative Writing',
              'Factual Accuracy', 'Multilingual', 'Context Length', 'Speed']
              
model_a = [95, 92, 90, 88, 85, 90, 95, 70]
model_b = [88, 95, 92, 80, 88, 85, 80, 85]
model_c = [82, 85, 85, 92, 80, 78, 70, 95]
model_d = [90, 88, 88, 85, 92, 92, 85, 80]

N = len(categories)
angles = np.linspace(0, 2 * np.pi, N, endpoint=False).tolist()
angles += angles[:1]

model_a += model_a[:1]
model_b += model_b[:1]
model_c += model_c[:1]
model_d += model_d[:1]

fig, ax = plt.subplots(figsize=(10, 10), subplot_kw=dict(polar=True), facecolor='#0a0a0f')
ax.set_facecolor('#0a0a0f')

ax.plot(angles, model_a, 'o-', linewidth=2.5, color='#5314E6', label='Claude', markersize=7)
ax.fill(angles, model_a, alpha=0.2, color='#5314E6')

ax.plot(angles, model_b, 's-', linewidth=2.5, color='#27D3F5', label='GPT-4', markersize=7)
ax.fill(angles, model_b, alpha=0.2, color='#27D3F5')

ax.plot(angles, model_c, '^-', linewidth=2.5, color='#6CF527', label='Gemini', markersize=7)
ax.fill(angles, model_c, alpha=0.2, color='#6CF527')

ax.plot(angles, model_d, 'D-', linewidth=2.5, color='#F5276C', label='Llama', markersize=7)
ax.fill(angles, model_d, alpha=0.2, color='#F5276C')

ax.set_xticks(angles[:-1])
ax.set_xticklabels(categories, fontsize=10, color='#e2e8f0', fontweight='500')
ax.set_ylim(0, 100)

ax.yaxis.grid(True, color='#1e293b', linestyle='-', linewidth=0.8)
ax.xaxis.grid(True, color='#334155', linestyle='-', linewidth=0.5)
ax.spines['polar'].set_color('#334155')
ax.tick_params(axis='y', colors='#94a3b8')

ax.set_title('AI Language Model Benchmarks', fontsize=16, color='#f8fafc',
             fontweight='bold', pad=25)

ax.legend(loc='upper right', bbox_to_anchor=(1.18, 1.1), fontsize=10,
          frameon=True, facecolor='#1e293b', edgecolor='#334155', labelcolor='#e2e8f0')

plt.tight_layout()
plt.show()