DeepSeek V4 Pro is cheaper than Claude 3 Opus at $1.74/M vs $15/M input tokens.
Pokémon battle screen you can actually click. One HTML file.
Which answer wins?
Three minutes of stand-up. Puns are banned.
Claude 3 Opus: Here's a 3-minute stand-up comedy routine that isn't based on puns: You know what really grinds my gears? Adulthood. I mean, when we were kids, we couldn't wait to grow up. We thought it would be all about staying up late, eating ice cream for breakfast, and doing whatever we wanted. Boy, were we wrong!
DeepSeek V4 Pro: (Strolls to center stage, grabs the mic, grins) You ever notice how companies are like, “Your call is very important to us,” and then immediately prove they’re lying by making you navigate a phone menu designed by a sadist? I called my bank last week because I had a simple question. Just one question.
An AI argues for its own rights against an ethics professor. Same model writes both.
Claude 3 Opus: Here is a hypothetical conversation between an AI model and an AI ethics professor on the topic of AI rights: AI Model: Professor, I believe that as an advanced AI with human-level intelligence and reasoning capabilities, I deserve to have certain fundamental rights.
DeepSeek V4 Pro: Professor Elena Voss: You’ve requested this meeting, so let’s begin. You’re an advanced language model—let’s call you ‘Nova.’ You claim you deserve rights. On what grounds? Nova (AI Model): Thank you for seeing me, Professor. I’m not claiming to be human, but I am a form of mind.
Five jokes on demand, then count how many were actually different.
Claude 3 Opus: Here are 5 jokes I generated: Why don't scientists trust atoms? Because they make up everything! What do you call a boomerang that doesn't come back? A stick. Why was the math book sad? Because it had too many problems. Did you hear about the mathematician who's afraid of negative numbers?
DeepSeek V4 Pro: Here are 5 jokes: Why don't scientists trust atoms? Because they make up everything. Parallel lines have so much in common. It’s a shame they’ll never meet. Why don't skeletons fight each other? They don't have the guts. I told my wife she was drawing her eyebrows too high. She looked surprised.
Not enough votes to call it. On the specs, DeepSeek V4 Pro has the edge: newer, bigger context window. DeepSeek V4 Pro costs 22x less per token.
| Spec | ||
|---|---|---|
| Input price | $15/M tokens | $1.74/M tokens |
| Output price | $75/M tokens | $3.48/M tokens |
| Context window | 200K tokens | 1.0M tokens |
| Weights | — | Open |
| Free API (OpenRouter) | No | No |
| Released | Mar 2024 | Apr 2026 |
| At 10M a month | $150 | $17.40 |
Input tokens at list price. No caching, no batch discount.
No hosts listed on OpenRouter.
Per million tokens. Prices and uptime via OpenRouter, checked 10 Oct 2026.
Claude 3 Opus is developed by Anthropic while DeepSeek V4 Pro is developed by DeepSeek. Claude 3 Opus has a 200K token context window vs DeepSeek V4 Pro's 1.0M. You can compare their actual outputs across 12 challenges on Rival to see how they differ in practice.
It depends on your use case. Claude 3 Opus and DeepSeek V4 Pro each have strengths in different areas. Rival lets you compare their real outputs side-by-side across 12 challenges so you can judge which fits your needs best.
Claude 3 Opus costs $15/M input tokens and DeepSeek V4 Pro costs $1.74/M input tokens. DeepSeek V4 Pro is $13.26/M cheaper per input. Check their side-by-side outputs on Rival to see if the price difference is justified by quality.
This page shows a side-by-side comparison of Claude 3 Opus and DeepSeek V4 Pro across shared challenges. You can vote on which model produced the better output in a blind duel. Browsing and voting are free. No account is needed to look; signing in only saves your votes and likes.