Language Model Comparisons
GPT, Claude, Gemini, Llama, and more — tested on real marketing, sales, code, and operations tasks.
Image Generator Comparisons
DALL-E, Midjourney alternatives, Stable Diffusion, Flux — tested on marketing, creative, and product imagery.
Code-Drawn Art: ASCII & SVG
Can a language model draw — not generate an image, but write the code for one? We test the same subjects two ways. The pattern is consistent and revealing: models are far better at SVG (structured coordinate code, their strength) than at ASCII (spatial layout in a character grid, their weakness) — and confidently claim success at both.
Return Only JSON
View benchmark →
ASCIIASCII Persian Cat
View benchmark →
ASCIIASCII Bar Chart
View benchmark →
ASCIIASCII Geometric Shapes
View benchmark →
ASCIIASCII Lion with Complex Shading
View benchmark →
ASCIIASCII Portrait - Einstein
View benchmark →
ASCIIGMTECH 90s Style ASCII Art
View benchmark →
Cross-Test Finding
Every model draws the same blob for a lion and a physicist
None of the ASCII portraits are actually recognizable — but the revealing part is how they fail. We measured the vertical silhouette of each model's lion against its Einstein portrait. A big cat and a human face should look nothing alike. They score 0.75–0.96 identical. Models don't render the subject — they emit one generic tapered shading blob and relabel it.
Lion-vs-Einstein silhouette similarity (vertical fill-ratio cosine, 1.0 = identical shape). Higher = the model reused the same shape for both subjects.
Run Your Own Comparisons in AI Lab
Test your actual prompts across every major model. See quality, cost, and speed side by side. Save what works. No commitment on the monthly plan.