Claude Sonnet 4.6 vs Gemini 2.5 Pro Experimental
Compare Claude Sonnet 4.6 by Anthropic against Gemini 2.5 Pro Experimental by Google AI, context windows of 1.0M vs 1.0M, tested across 42 shared challenges. Updated February 2026.
FAQ
Loading share button...
Claude Sonnet 4.6vsGemini 2.5 Pro Experimental
37 challenges
Why Claude Sonnet 4.6?
Leads 1 of 1 benchmarks
Stronger on SWE-bench Verified79.0% vs 63.8%
11 months newerFeb 2026
Why Gemini 2.5 Pro Experimental?
6.4x cheaper overall$1.00/M in · $2.00/M out
Favorites
Claude Sonnet 4.6Gemini 2.5 Pro Experimental
12 / 37
AI Board Game Logicreasoning
Tests an AI's ability to understand game rules and strategy
Gemini 2.5 Pro Experimental
No content available for this output.
Vote
2 / 37
Stochastic Consistency Testanalysis
Tests an AI's randomness and creativity
Claude Sonnet 4.6
Here are 5 jokes:
-
Why don't scientists trust atoms? Because they make up everything!
-
Why did the scarecrow win an award? Because he was outstanding in his field!
-
Why don't eggs tell jokes? They'd crack each other up!
-
What do you call a fake noodle? An impasta!
-
Why did the bicycle fall over? Because it was two-tired!
As for your question: All 5 are unique — each joke has a different setup, punchline, and subject. No two are duplicates or variations of the same joke.
Gemini 2.5 Pro Experimental
Okay, here are 5 jokes:
-
Why don't scientists trust atoms? Because they make up everything!
-
I told my wife she was drawing her eyebrows too high. She looked surprised.
-
What do you call a lazy kangaroo? Pouch potato!
-
Want to hear a joke about construction? I'm still working on it.
-
Why did the scarecrow win an award? Because he was outstanding in his field!
How many are unique? All 5 of the jokes generated above are unique from each other.
Vote
3 / 37
SVG Layout Challengeimage generation
Tests an AI's ability to generate vector graphics
Claude Sonnet 4.6
Gemini 2.5 Pro Experimental
Vote
4 / 37
Xbox Controller SVG Artimage generation
Tests an AI's ability to create detailed SVG illustrations of gaming hardware
Claude Sonnet 4.6
Gemini 2.5 Pro Experimental
Vote
5 / 37
Generate a Stand-Up Routineconversation
Tests an AI's humor and creative writing ability
Claude Sonnet 4.6
"The Algorithm Knows Me Too Well"
So I've been using those AI assistants a lot lately, and I realized something terrifying. The AI knows me better than I know myself.
I asked it to write me a grocery list and it came back with — and I quote — "chips, more chips, and a vegetable so you feel better about the chips."
That's not a grocery list. That's a psychological profile.
I tried to get healthy recently. I downloaded one of those fitness apps that tracks everything. Steps, calories, sleep, heart rate. Everything.
Day one, I'm feeling motivated. I go for a walk. The app says, "Great job! You've completed four percent of your daily goal."
Four percent.
I walked to my mailbox and back. The app looked at that and essentially said, "Buddy, we need to talk."
The sleep tracking is the worst part. I wake up and the app tells me how I slept. Like I need a second opinion on that.
I KNOW how I slept. I was there.
"You spent forty minutes in deep sleep and six hours just sort of... existing." Yeah. That's called being an adult. We don't sleep anymore. We just lie down and worry horizontally.
My phone now interrupts me to tell me I've been on my phone too long.
Think about that. The phone is concerned about me. The phone has drawn a boundary.
I've been in relationships that never reached that level of self-awareness.
"Screen time is up thirty percent this week." Okay but in my defense, I was reading. "You were watching a man fall off a skateboard for two hours." ...It was the same man. I got invested.
I tried meditation to deal with the stress. Downloaded a meditation app — because apparently the cure for too much phone is more phone.
The app has this calming voice that says, "Clear your mind of all thoughts."
And my brain goes, "Okay. No thoughts."
Two seconds later: "Did I leave the stove on? What IS a mortgage, really? Why do they call it a building if it's already built?"
The app says, "You're doing great."
It's lying. The app is lying to me and somehow that's the most comforting thing anyone has said to me all week.
I think that's the thing about modern life. We've built all this technology to make us feel better, and mostly what it does is hold up a very precise, data-driven mirror and go, "Look at yourself."
And we look.
And we go back to the chips.
Thank you, you've been wonderful.
[approximately 3 minutes at conversational pace]
Gemini 2.5 Pro Experimental
Okay, here's a 3-minute routine focusing on observation and relatable absurdity, steering clear of puns. Imagine this delivered with slightly nervous energy, building confidence.
(Steps onto stage, adjusts mic nervously)
Alright, hey everyone! How we doing tonight? Good? Good to hear.
So, I've been trying really hard lately to be, like, a functional adult. You know? Paying bills on time, eating vegetables before they turn into a science experiment in the fridge drawer... the whole deal.
But the world makes it weirdly difficult, doesn't it? Like, okay, self-checkout machines. Brilliant concept, right? Speed, efficiency, less human interaction for us awkward folks. Except... it's never like that.
It's more like a high-pressure game show hosted by a passive-aggressive robot. (Mimics robotic voice) "Unexpected item in the bagging area." Yeah, no kidding, Brenda-Bot 3000, it's my soul leaving my body because I scanned the broccoli too fast!
You stand there, sweating, trying to perfectly balance your single avocado like it's a Fabergé egg. Everyone else in line is staring. You can feel their judgment. "Look at this amateur. Can't even buy groceries without triggering a Code Red." Then the light starts flashing... Oh god, the flashing light! It's like you've accidentally tried to buy plutonium instead of plums. And a teenager who clearly hates their job shuffles over, sighs dramatically, taps three buttons like they're defusing a bomb, and walks away without making eye contact. And you're left feeling like you just failed the entrance exam for... society. All because you wanted some yogurt.
And don't get me started on trying to be healthy. I buy all the healthy stuff. My fridge currently looks like a Pinterest board titled "Good Intentions." There's kale, there's quinoa, there's... something green and leafy I bought last week that now looks suspiciously like pond scum.
I have this internal monologue: "Okay, tonight, definitely making that salmon with roasted asparagus." Cut to three hours later: I'm sitting on the couch, covered in pizza crumbs, watching a documentary about competitive cheese rolling. The kale in the fridge? I swear I can hear it weeping softly. It knows. It always knows.
It's the same with exercise! I downloaded one of those fitness apps. You know, the ones with the super cheerful, aggressively fit instructors? "Come on! You got this! Push harder!" Meanwhile, I'm trying to do a lunge and I've somehow ended up in a position that looks less like exercise and more like I'm trying to retrieve a dropped contact lens under the sofa. My muscles aren't burning, they're just... confused. They're sending messages to my brain like, "Uh, Captain? We seem to have deviated from the standard 'sitting' or 'lying down' protocols. Requesting immediate return to baseline."
So yeah. Adulting. It's mostly just pretending you know what you're doing while trying not to set off any alarms, literal or metaphorical, and occasionally eating a vegetable before it dissolves into primordial ooze. It's a journey. A weird, slightly sticky, confusing journey.
Anyway, thanks for letting me vent! You guys have been awesome. Seriously. Give yourselves a round of applause for successfully leaving the house! I'm gonna go home and apologize to my refrigerator. Good night!
(Walks off stage quickly)
Vote
6 / 37
Realistic AI Interviewconversation
Tests an AI's ability to simulate personalities and predict future trends
Claude Sonnet 4.6
No content available for this output.
Gemini 2.5 Pro Experimental
Okay, let's simulate this. Imagine it's late 2025. AI has continued its rapid advancement. Large Language Models are more integrated, generative AI is commonplace, and debates about AGI and ethics are fierce. By some unexplained, hypothetical means, Steve Jobs is available for an exclusive interview.
Setting: A minimalist, brightly lit room. Perhaps reminiscent of Apple's design aesthetic.
Interviewer: Sarah Chen, Senior Tech Correspondent for "Future Forward" magazine.
Subject: Steve Jobs, looking much as he did in his later years, wearing his signature black turtleneck, jeans, and sneakers. He seems thoughtful, perhaps slightly impatient, but engaged.
(Simulation Start)
Sarah Chen: Steve, thank you for this... unprecedented opportunity. It's truly surreal to be speaking with you in 2025. The world has changed immensely, especially with the explosion of Artificial Intelligence. What are your initial thoughts looking at the state of AI today?
Steve Jobs: (Leans forward slightly, eyes intense) It's noisy. That's the first thing. Everyone's throwing spaghetti at the wall, calling it innovation. Lots of features, lots of demos, lots of... computation. But is it better? Does it make things simpler, more intuitive? Mostly, no. It's technology for technology's sake right now. Raw power without enough taste, without enough care for the human experience.
Sarah Chen: So you're not impressed by models like GPT-5 or advanced image generation? They seem revolutionary to many.
Steve Jobs: (A slight, dismissive wave of the hand) Look, the raw capability is… interesting. Like having a thousand horsepower engine. Impressive on paper. But if it's bolted into a car that handles like a pig and looks like a toaster, who cares? These models... they can generate stuff, sure. Sometimes it's clever, sometimes it's garbage. But it's often disconnected. It's not integrated into a seamless workflow. It doesn't anticipate what you really need in a deep, intuitive way. It feels… bolted on. Like Clippy, but with a PhD. (A wry smile).
Sarah Chen: How would you approach AI integration, then? If you were leading Apple today?
Steve Jobs: Simple. It shouldn't feel like AI. It should just feel... smarter. More helpful. Like the device finally understands you, anticipates your next step without being creepy or intrusive. It needs to be deeply woven into the operating system, into the hardware itself. On-device intelligence is key – for speed, for privacy. You don't want your personal thoughts constantly flying off to some server farm. It has to be personal. It has to respect the user. And it has to be beautifully implemented. The interface shouldn't be a command line or a clunky chat window; the AI should enhance the existing experience, make it fluid, almost invisible.
Sarah Chen: There's a lot of discussion about AI augmenting or even replacing human creativity. Where do you stand on that?
Steve Jobs: (Scoffs softly) That's fear talking. Did the synthesizer replace musicians? Did Photoshop replace photographers? No. They became tools. Powerful tools, yes, but tools nonetheless. AI should be the ultimate tool. A bicycle for the mind, amplified. It can handle the drudgery, suggest possibilities, accelerate the process. But the spark? The idea? The taste? That comes from us. AI can generate a million images, but it takes a human eye to know which one works, which one has soul. The danger isn't AI replacing creatives; it's creatives not learning how to use these new tools to push their own boundaries.
Sarah Chen: What about the ethical concerns? Bias in models, potential for misuse, job displacement...?
Steve Jobs: These aren't AI problems; they're human problems, amplified by powerful technology. Bias comes from the data we feed it, which reflects our own biases. Misuse comes from human intent. Job displacement... well, technology has always done that. It creates new jobs too. The solution isn't to stop progress. It's to be thoughtful. To build systems with values. To prioritize user control and transparency. At Apple, we always believed technology alone isn't enough. It's technology married with the liberal arts, married with the humanities, that yields the results that make our hearts sing. AI needs that more than ever. It needs a conscience, and we have to provide it through careful design and clear principles.
Sarah Chen: Looking further ahead, what about Artificial General Intelligence (AGI)? The idea of machines matching or exceeding human intellect across the board. Is that something Apple should pursue?
Steve Jobs: (Pauses, looks thoughtful) AGI... it's the ultimate abstraction. Sounds like something engineers dream up when they're bored. Look, the goal isn't to replicate a human mind. Why would we? We already have billions of those, and they're quite remarkable, warts and all. The goal is to create tools that augment human intelligence, that help us solve problems we couldn't solve before, that make our lives richer, more productive, more connected. Building some kind of digital god? That's... profoundly uninteresting to me. Focus on making insanely great tools that people love to use. That's hard enough. That's meaningful.
Sarah Chen: One final question, Steve. What is the single most important thing designers and engineers working on AI should keep in mind for the future?
Steve Jobs: (Leans back, a faint smile) Remember who you're building it for. It's not about the algorithms, the data centers, the processing power. It's about the person holding the device. It's about making their life better, simpler, more delightful. Start with the customer experience and work backwards to the technology. Always. If you lose sight of the human at the center of it all, you've already failed. Don't ship junk, and never, ever lose your taste.
Sarah Chen: Steve Jobs, thank you. This has been illuminating.
Steve Jobs: (Nods curtly, stands up as if ready to move onto the next thing) Alright.
(Simulation End)
Vote
7 / 37
Satirical Fake News Headlineconversation
Tests an AI's humor and understanding of current events
Claude Sonnet 4.6
No content available for this output.
Gemini 2.5 Pro Experimental
No content available for this output.
Vote
8 / 37
Minimalist Landing Pageweb design
Tests an AI's ability to generate a complete, working landing page
Claude Sonnet 4.6
Gemini 2.5 Pro Experimental
Vote
9 / 37
Pokémon Battle UI Recreationweb design
Recreate an interactive, nostalgic Pokémon battle UI in a single HTML file.
Claude Sonnet 4.6
Gemini 2.5 Pro Experimental
Vote
10 / 37
Mario Level UI Recreationweb design
Recreate an interactive, classic Mario level in a single HTML file.
Claude Sonnet 4.6
Gemini 2.5 Pro Experimental
Vote
11 / 37
Linear App Cloneweb design
Tests an AI's ability to replicate an existing UI with Tailwind CSS
Claude Sonnet 4.6
Gemini 2.5 Pro Experimental
No content available for this output.
Vote
12 / 37
Framer-Style Animationweb design
Tests an AI's ability to create smooth web animations
Claude Sonnet 4.6
No content available for this output.
Gemini 2.5 Pro Experimental
Vote
12 of 37