AI Content Benchmark

Compare 4 AI writing
methods in real-time

Pick a service, click Launch — watch GPT-4.1 web research, GPT-5 Mini,
Gemini 2.5 Flash and Claude Haiku generate live, side by side.

4 Generation Methods
🔍
V1 — Consumer Research
gpt-4.1 web search → gpt-5-mini
4 live web searches for pricing, competitors, reviews and company data — then generates a 2,000+ word guide.
~2min€0.10/article10/10 quality
⚡
V2 — GPT-5 Mini
gpt-5-mini-2025-08-07
Direct generation using the model's training knowledge. Fast, cost-efficient, surprisingly thorough.
~90sec€0.011/article9/10 quality
✨
V3 — Gemini 2.5 Flash
gemini-2.5-flash
Google's fast multimodal model. Strong reasoning, great HTML structure, highly cost-efficient.
~20sec€0.003/article8/10 quality
🤖
V4 — Claude Haiku 4.5
claude-haiku-4-5-20251001
Anthropic's fastest frontier model. Near-frontier intelligence, precise instructions, clean prose.
~25sec€0.005/article9/10 quality
Choose a Service
No service selected
Initialising…
🔍
V1 — Consumer Research
gpt-4.1 + gpt-5-mini
waiting
Content will appear here…
⚡
V2 — GPT-5 Mini
gpt-5-mini direct
waiting
Content will appear here…
✨
V3 — Gemini Flash
gemini-2.5-flash
waiting
Content will appear here…
🤖
V4 — Claude Haiku
claude-haiku-4-5
waiting
Content will appear here…

Generation Complete

Click any card to read the full article. Compare them side by side below.

Side-by-Side Comparison
Test another service? Run the benchmark again with a different service or method.