Pick a task. Choose a model from Anthropic or OpenAI, run it, and see the head-to-head.
VS
Required Capabilities
Which Models Excel?
Select an activity
SimpleAnthropicSonnet 5Medium
Response Time
0
seconds
Output Quality
0
out of 100
Good
Speed × Quality
0
composite
Tokens & Cost
0
$0.00
Example output
Click Run to see results
⚖️ Head-to-head
VS
Share Your Results
Twitter
LinkedIn
Facebook
WhatsApp
Telegram
Instagram
Reddit
Copy
⚙️ Your Settings
Difficulty:Simple
Activity:—
Provider:Anthropic
Model:Sonnet 5
Compared with:—
Effort:Medium
Extended:Off
Composite:—
Metrics are synthesized from recorded-run profiles (no live calls). OpenAI profiles are placeholders until runs are recorded. Extended Thinking is a Claude feature and stays off for OpenAI models. Token cost uses each model's published per-million-token price. How we measure → · FAQ → · Model comparison →
Welcome! What should we call you?
We'll use your name instead of "You" on your results — makes it feel like your own race.
🏁
Here's how it works
Pick a task you already do every week — like writing an email.
Pick an AI model — Claude, GPT, whichever you're curious about.
Watch the race: how long you'd take, next to how long the AI took.
💡
What you'll discover
Real speed gaps between you and AI, task by task.
How accurate each answer actually is — speed isn't the whole story.
The surprise: the same task can cost 10x more depending on which model you pick.