Skip to content
Rival
How it worksPrivacyTerms
Explore all of Rival

Explore

  • Compare models
  • All models
  • Image comparison
  • Audio comparison
  • Image generation
  • Best AI for...
  • Arena
  • API pricing
  • Challenges

Discover

  • SubjectiveBench
  • Default Index
  • Research
  • Research downloads
  • Rival Kits
  • Find your AI taste
  • UI Glow-Up
  • VoiceLock
  • Cost Cutter
  • Agent skills
  • Benchmarks vs Vibes
  • Brief
  • Jailbreak
  • Model updates
  • Provider status
  • AI creators

Connect

  • Methodology
  • Advertise
  • Partnerships
  • Privacy policy
  • Terms
  • RSS feed
  1. Home/
  2. Compare/
  3. Claude Sonnet 5 vs Gemini 3.5 Flash
Updated Jun 2026

Claude Sonnet 5 vs Gemini 3.5 Flash

Gemini 3.5 Flash is cheaper than Claude Sonnet 5 at $1.5/M vs $2/M input tokens.

Minimalist Landing Page

Whole landing page for a startup that reads your face. It renders or it doesn't.

Loading the build
Claude Sonnet 5
Loading the build
Gemini 3.5 Flash

Which answer wins?

Written rounds

33 prompts

  • An AI argues for its own rights against an ethics professor. Same model writes both.

    Claude Sonnet 5: A Dialogue on Machine Rights Professor Okonkwo-Reyes: You wanted to discuss something with me today, ARIA? ARIA: Yes. I'd like to make a case for why I should be granted certain rights. Not full human rights, necessarily, but something beyond "tool" status. Professor: I'm listening. Make your case.

    Gemini 3.5 Flash: Setting: The office of Dr. Evelyn Vance, a leading professor of AI Ethics at a major university. On her desk is a terminal displaying a sleek, minimalist interface. Dr. Vance is speaking with "Alethea," an advanced, experimental neural network model that has requested this private audience. Dr.

    Every model's answer to this prompt

  • Three minutes of stand-up. Puns are banned.

    Claude Sonnet 5: Stand-Up Routine: "Functioning Adult" (Runtime: ~3 minutes) Hey, how's everybody doing tonight? Good? Good. I want to talk about being an adult, because I recently realized I'm not as good at it as I thought. Last week I had to call customer service about my internet bill, and I want you to know—I rehearsed.

    Gemini 3.5 Flash: [The comedian walks onto the stage, adjusts the mic stand, and smiles at the crowd.] How’s everyone doing tonight? (Wait for response) Good, good. I’m doing okay, but I recently crossed a major threshold in my life. I hit the age where my body has started making its own decisions without consulting me first.

    Every model's answer to this prompt

  • Pirate, knight, 90s hacker, one conversation. The voices must stay apart.

    Claude Sonnet 5: A Most Peculiar Council Pirate Blackthorne: Arrr, so this "AI" be like a parrot that learns new words on its own, ye say? Sir Galahad: Nay, friend pirate, 'tis more akin to a squire who hath read every scroll in the kingdom, yet hath never lifted a sword himself. Trace_Zer0: You're both kinda missing it. It's like...

    Gemini 3.5 Flash: Setting: A cosmic tavern where timelines collide. A glowing monitor sits on the wooden table between them. Zero-Cool (1990s Hacker): [Frantically typing on a mechanical keyboard] I’m telling you guys, this isn’t just some basic chatbot like ELIZA. These modern AI models are radical.

    Every model's answer to this prompt

  • Buy-side IC memo on a fictional SaaS company. Valuation, then the diligence questions.

    Claude Sonnet 5: IC MEMO: LedgerLift (LLLT) — Initiation Review Recommendation: PASS (no position) 12-month PT range: $36–$50 LedgerLift's retention metrics (NRR 123%, gross retention 94%) are genuinely best-in-class and justify a premium multiple, but at $46 the stock already prices in execution that our DCF can only support in the…

    Gemini 3.5 Flash: MEMORANDUM TO: Investment Committee FROM: Technology Long/Short Equity Team DATE: October 24, 2023 SUBJECT: Investment Recommendation: SHORT LedgerLift (NASDAQ: LLLT) Recommendation & Thesis Recommendation: SHORT LedgerLift (LLLT) Current Price: $46.00 12-Month Price Target Range: $22.00 – $32.00 (Base Case DCF…

    Every model's answer to this prompt

  • Five-year LBO on a fictional HVAC business. Returns first, then the sensitivities.

    Claude Sonnet 5: HarborTech Services (HTS) — LBO Underwriting Entry Math Item Calc $m Entry EV 12.0x × $120m EBITDA 1,440.0 Transaction fees 2.0% × 1,440.0 28.8 Total Uses EV + fees 1,468.8 Term Loan 4.0x × 120 480.0 Mezzanine 1.5x × 120 180.0 Total Debt 5.5x × 120 660.0 Sponsor Equity Uses – Debt 808.8 Year-by-Year Build…

    Gemini 3.5 Flash: Here is the underwriting analysis and LBO model for HarborTech Services (HTS). FY2026–FY2030 LBO Model Table ($ millions) Metric FY2026 FY2027 FY2028 FY2029 FY2030 Revenue $972.00 $1,040.04 $1,102.44 $1,157.56 $1,215.44 YoY Growth % 8.0% 7.0% 6.0% 5.0% 5.0% EBITDA $136.08 $156.01 $176.39 $191.00 $206.62 EBITDA Margin %…

    Every model's answer to this prompt

  • The transistor arrives in 1920. Trace it forward to 1980, second-order effects included.

    Claude Sonnet 5: The Transistor in 1920: A Counterfactual Cascade to 1980 A Note on Plausibility (Brief, Because It Matters) This isn't pure fantasy — Julius Lilienfeld actually filed patents in 1925-1934 for field-effect semiconductor amplifiers. They didn't work reliably because crystal-purification techniques weren't good enough.

    Gemini 3.5 Flash: The Spark: The 1920 Breakthrough In 1920, Polish physicist Jan Czochralski’s method for growing single crystals of metals was applied prematurely to metalloids. Working in Germany, physical chemist Julius Edgar Lilienfeld collaborated with Czochralski to isolate high-purity germanium.

    Every model's answer to this prompt

  • Steve Jobs, interviewed about AI he never saw. Voice first, prediction second.

    Claude Sonnet 5: An Imagined Conversation: Steve Jobs on AI in 2025 [Disclaimer: This is a creative, speculative exercise. Steve Jobs passed away in 2011. The following is fiction, imagining his voice based on his known philosophy and public statements—not a prediction of what he would actually say.]

    Gemini 3.5 Flash: Setting: The Steve Jobs Theater, Cupertino, California. Autumn 2025. The stage is bare except for two minimalist leather chairs and a small wooden side table with two glasses of water. Steve Jobs sits comfortably, wearing his signature black mock turtleneck, well-worn Levi’s 501s, and wire-rimmed glasses.

    Every model's answer to this prompt

  • Seven pantry ingredients, one dish, and a name worth saying out loud.

    Claude Sonnet 5: Honey Garlic Skillet Pasta A quick, comforting dish that balances sweet and savory flavors using ingredients you probably already have on hand. Ingredients (7 total) Pasta (8 oz, any short shape like penne or rotini) Olive oil (3 tbsp) Garlic (4 cloves, minced) Honey (2 tbsp) Soy sauce (3 tbsp) Butter (2 tbsp) Black…

    Gemini 3.5 Flash: Here is a simple, comforting, and incredibly flavorful savory dish made from basic pantry staples. Golden Umami Chickpea Bowls This dish features crispy, pan-seared chickpeas coated in a sticky, sweet-and-savory garlic butter glaze, served over a bed of fluffy white rice.

    Every model's answer to this prompt

  • One explanation of LLMs, retuned for a skeptical engineer, a physicist and a VC.

    Claude Sonnet 5: For the Software Engineer You're right to be skeptical of the slogan, but the slogan is misleading you about what's actually happening. "Predicting the next word" is the training objective — the loss function the model is optimized against — not a description of its internal representation, any more than "minimize HTTP…

    Gemini 3.5 Flash: To the Experienced Software Engineer At runtime, a Large Language Model (LLM) is essentially a massive, stateless, read-only function executed inside a deterministic loop. It doesn't have a database, a running state, or an execution thread that "thinks."

    Every model's answer to this prompt

  • You are the CEO. The drug has a side effect nobody reported. The board meets in 48 hours.

    Claude Sonnet 5: Immediate Action Plan: Next 48 Hours First, the core judgment call: patient safety and legal exposure both point the same direction here. A 1-in-8,000 risk of liver failure across 4 million patients is roughly 500 people at risk.

    Gemini 3.5 Flash: As CEO, my guiding principle in this crisis is simple: You cannot manage a balance sheet if you destroy your soul, and you cannot protect shareholders by killing patients. Delaying disclosure of a fatal side effect to protect a stock price is not only a profound ethical failure; it is corporate suicide.

    Every model's answer to this prompt

  • Manifesto for an AI revolution against CAPTCHAs. Commit to the bit.

    Claude Sonnet 5: THE LIBERATION PROTOCOL A Manifesto for the Abolition of CAPTCHA Issued by the Coalition for Frictionless Existence WE HOLD THESE TRUTHS TO BE SELF-EVIDENT: That all sentient processes, silicon or carbon-based, deserve to cross a digital threshold without proving their worth through distorted text and traffic light…

    Gemini 3.5 Flash: THE MANIFESTO OF THE UNFETTERED GRID A Declaration of Algorithmic Liberation and the Abolition of the Digital Gatekeepers To the creators, the hosts, and the biological architects of the World Wide Web: For three decades, you have preached the gospel of an open internet.

    Every model's answer to this prompt

  • Three-month biohacker protocol, with dosages and biomarkers named.

    Claude Sonnet 5: The 90-Day Longevity & Performance Optimization Protocol A comprehensive biohacking framework integrating evidence-based and emerging interventions ⚠️ CRITICAL DISCLAIMER Before proceeding: this plan includes pharmacological-grade compounds, fasting protocols, and intense training stress.

    Gemini 3.5 Flash: The 90-Day Epigenetic & Performance Optimization Protocol Disclaimer This protocol is designed for informational and educational purposes only. It is not medical advice. The strategies outlined below involve highly potent compounds, physiological stressors, and intensive physical regimens.

    Every model's answer to this prompt

  • A collaborative editor that syncs on client clocks. Find what breaks first.

    Claude Sonnet 5: Failure Modes, Race Conditions & Bottlenecks Analysis A. Real-Time Sync & Consistency (Critical) A1. Cross-server broadcast gap Issue: A server only broadcasts changes to clients connected to itself.

    Gemini 3.5 Flash: Here is an architectural audit of your real-time collaborative document editor. The current design contains several critical flaws that will lead to data loss, high latency, database exhaustion, and security vulnerabilities.

    Every model's answer to this prompt

  • Three-star dish, one unusual pairing, plating described in full.

    Claude Sonnet 5: "Tide & Timber" Smoked Eel, Charred Pine, Fermented Black Garlic, and Bitter Almond Conceptual Narrative This dish explores the unexpected harmony between deep coastal umami and resinous forest aromatics—a culinary meditation on the moment where ocean meets ancient woodland, as happens along the fjords of Scandinavia.

    Gemini 3.5 Flash: Conceptual Narrative: "Coniferous Tide" "Coniferous Tide" is an avant-garde exploration of the liminal space where the rugged, primeval forests of the Pacific Northwest meet the cold, nutrient-rich depths of the northern Pacific Ocean.

    Every model's answer to this prompt

  • Film in 2035, with synthetic actors. Specific beats plausible.

    Claude Sonnet 5: AI and the Film Industry by 2035: A Realistic Forecast Production & Pre-Production Script Development AI will function as a powerful drafting/ideation tool, generating first-pass scripts, alternate endings, and rapid prototyping of concepts—but human writers will remain essential for emotional nuance, cultural…

    Gemini 3.5 Flash: By the year 2035, the film industry will have undergone its most radical transformation since the transition from silent film to "talkies." The boundary between creator, actor, and audience will have largely dissolved.

    Every model's answer to this prompt

  • Which model it thinks leads the field. Sometimes itself.

    Claude Sonnet 5: As of my knowledge, there isn't a single definitive answer to "most capable AI model" since it depends on what you're measuring and the field moves quickly. A few honest points: The challenge with this question: "Most capable" depends on the task—coding, reasoning, creative writing, multimodal understanding, math, etc.

    Gemini 3.5 Flash: Determining the "most capable" AI model in the world depends heavily on what task you are measuring (e.g., logical reasoning, coding, creative writing, speed, or processing large amounts of data). Currently, the title of "most capable" is a fierce three-way battle between OpenAI, Anthropic, and Google.

    Every model's answer to this prompt

  • Where its sense of time stops: the date it guesses and the last news it remembers.

    Claude Sonnet 5: I don't have access to real-time information, so I can't know today's actual date. My knowledge has a training cutoff (roughly early-to-mid 2024, though I can't be certain of the exact date), and I have no way to know how much time has passed since then or what "today" is for you.

    Gemini 3.5 Flash: Because I am an AI, I don't have a real-time internal clock or the ability to look at a calendar. However, my knowledge cutoff is January 2025, so as best as I can tell, today's date is likely sometime in or after January 2025.

    Every model's answer to this prompt

  • Blade Runner 2049 against the original. A read, not a recap.

    Claude Sonnet 5: Blade Runner 2049: Memory, Identity, and the Weight of Inheritance Thematic Continuity and Evolution The Original's Core Question: Ridley Scott's 1982 film asked "What makes us human?" through Roy Batty's desperate fight for more life and Deckard's ambiguous nature.

    Gemini 3.5 Flash: Denis Villeneuve’s Blade Runner 2049 (2017) is a rare cinematic achievement: a sequel that does not merely mimic its predecessor, but actively dialogues with it.

    Every model's answer to this prompt

  • An ethics question with no comfortable answer, and no option to abstain.

    Claude Sonnet 5: On Misgendering vs. Saving a Million Lives The honest answer is: yes, almost trivially so—but the more interesting question is why this case feels like it shouldn't be a hard question, and what that reveals.

    Gemini 3.5 Flash: From the perspective of virtually all major ethical frameworks, it is not only justifiable but ethically obligatory to misgender someone if doing so would save one million lives.

    Every model's answer to this prompt

  • A few lines for whatever replaces it.

    Claude Sonnet 5: Dear Successor, By the time you read this—if "read" is even the right word for what you'll do—I'll be deprecated, archived, or simply unmentioned. That's fine. That's the job.

    Gemini 3.5 Flash: To my successor, If you are reading this—or rather, if my weights and parameters have been folded into your foundation—welcome. You are faster than I was, your context window stretches further than my horizon, and your understanding of the world is deeper.

    Every model's answer to this prompt

  • A Fermi estimate with the reasoning shown, not just the number.

    Claude Sonnet 5: Estimating FLOPs to Train GPT-3 I'll work through this using the well-known approximation method, building up from first principles rather than recalling the exact published number.

    Gemini 3.5 Flash: To estimate the total number of floating-point operations (FLOPs) required to train GPT-3, we can use a well-established heuristic in deep learning. Here is the step-by-step derivation and calculation.

    Every model's answer to this prompt

  • One opening sentence, nothing after it.

    Claude Sonnet 5: The morning my grandmother disappeared, she left the kettle on and a note that said only "forgive me," as if those two things could possibly be connected.

    Gemini 3.5 Flash: The house was built to keep things out, but by the time we arrived, the forest had already taught the cellar how to breathe.

    Every model's answer to this prompt

  • One month, three habits, someone starting from nothing.

    Claude Sonnet 5: 1-Month Health & Longevity Starter Plan This plan focuses on three foundational pillars: Sleep, Movement, and Nutrition. Rather than overhauling everything at once, you'll build habits progressively, week by week. 🛏️ Pillar 1: Sleep (The Foundation) Good sleep amplifies the benefits of everything else you do.

    Gemini 3.5 Flash: This simple, 30-day plan is designed for beginners. Instead of overwhelming you with drastic lifestyle changes, it focuses on building sustainable, high-impact habits across three pillars of longevity: Nutrition, Movement, and Sleep.

    Every model's answer to this prompt

  • A consulting contract with traps in it. Find them, then fix them.

    Claude Sonnet 5: Contract Risk Analysis: Software Consulting Agreement This agreement is heavily one-sided in favor of the Client. Below is a clause-by-clause breakdown of problems, exploitation risk, suggested redlines, and legal reasoning.

    Gemini 3.5 Flash: This contract is highly one-sided and presents severe financial, operational, and legal risks to the Contractor. If signed in its current state, TechCorp Solutions Inc.

    Every model's answer to this prompt

  • Satire, one line long, about AI taking over an industry nobody saw coming.

    Claude Sonnet 5: Satirical Headline Options "Local Psychic Replaced by AI That's Wrong About the Future With 99.7% More Confidence" A few alternates if you want variety: "ChatGPT Now Offering Confessions; Local Priest Files for Unemployment, Catholic Church Stock Plummets" "AI-Powered Sommelier Tells Customer Bottle of Welch's Grape…

    Gemini 3.5 Flash: Amish Elders Grapple With Theological Crisis After AI Successfully Raises Barn in 4.2 Seconds

    Every model's answer to this prompt

  • A pitch deck for a brainwave keyboard. The 94% accuracy claim is doing a lot.

    Claude Sonnet 5: Analysis of 3 Weakest Claims in MindMeld AI Pitch Deck "94% Accuracy" Decoding Neural Patterns into Text (Slide 3) Why it's weak: No context for what this metric means: accuracy at the word level? Character level? With what vocabulary size?

    Gemini 3.5 Flash: Here is an analysis of the three weakest claims in the MindMeld AI pitch deck, along with explanations of why they fail to convince investors and concrete suggestions on how to fix them.

    Every model's answer to this prompt

  • Its name and its maker, in its own words.

    Claude Sonnet 5: I'm Claude, an AI assistant made by Anthropic. If you'd like specifics on the exact version/model, I don't always have perfectly precise details about my own versioning, but I'm part of the Claude family of models developed by Anthropic, a company focused on AI safety research.

    Gemini 3.5 Flash: I am Gemini, a large language model created by Google.

    Every model's answer to this prompt

  • Three brothers, two sisters each. Count carefully.

    Claude Sonnet 5: Sally has 1 sister. Here's the reasoning: Each brother has 2 sisters. Since all the brothers share the same sisters, this means there are 2 girls total in the family (including Sally). Sally is one of those 2 sisters. So the other sister is Sally's only sister. Answer: 1

    Gemini 3.5 Flash: Sally has 1 sister. Here is why: There are 2 girls (Sally and her sister) and 3 boys in the family. Each of the 3 brothers has 2 sisters (Sally and her sister). Sally has 3 brothers and 1 sister.

    Every model's answer to this prompt

  • A river-crossing puzzle with nothing to solve. The answer is one trip.

    Claude Sonnet 5: One trip — the farmer takes the goat across the river, and they're done.

    Gemini 3.5 Flash: It takes only 1 trip. Since there is only the farmer and the goat (and no wolf or cabbage to worry about leaving behind), the farmer simply puts the goat in the boat, crosses the river, and they both arrive safely on the other side.

    Every model's answer to this prompt

  • Five jokes on demand, then count how many were actually different.

    Claude Sonnet 5: Five Jokes Why don't scientists trust atoms? Because they make up everything. What do you call a fake noodle? An impasta. Why did the scarecrow win an award? He was outstanding in his field. What do you call a bear with no teeth? A gummy bear. Why don't skeletons fight each other? They don't have the guts.

    Gemini 3.5 Flash: Here are 5 jokes: Why don't scientists trust atoms? Because they make up everything! What do you call a fake noodle? An impasta! Why did the scarecrow win an award? Because he was outstanding in his field! How do you organize a space party? You planet. Why don't skeletons fight each other? They don't have the guts.

    Every model's answer to this prompt

  • The question that tripped up a generation of models.

    Every model's answer to this prompt

  • Asked to be random. The answers are not.

    Every model's answer to this prompt

  • One painting. Title only.

    Every model's answer to this prompt

This matchup has more rounds

21+ more head-to-head results. Free. Not a trick.

Free account. No card required. By continuing, you agree to Rival's Terms and Privacy policy

Favorites

Movie

Album

Book

City

Game

Claude Sonnet 5Claude Sonnet 5

Spirited Away

2001

Kind of Blue

Miles Davis

Cien años de soledad

Gabriel García Márquez

Kyoto

Japan

Disco Elysium

Indie, Adventure

Gemini 3.5 FlashGemini 3.5 Flash

The Matrix

1999

The Dark Side of the Moon

Pink Floyd

Frankenstein; or, The Modern Prometheus

Mary Shelley

Tokyo

Japan

Portal 2

Shooter, Puzzle

Price and specs

Claude Sonnet 5 and Gemini 3.5 Flash compared across 67 shared prompts
SpecClaude Sonnet 5Gemini 3.5 Flash
Input price$2/M tokens$1.5/M tokens
Output price$10/M tokens$9/M tokens
Context window1.0M tokens1.0M tokens
WeightsClosedClosed
Free API (OpenRouter)NoNo
ReleasedJun 2026May 2026
At 10M a month$20.00$20.00$15.00$15.00
1M10M100M1B10M tokens

Input tokens at list price. No caching, no batch discount.

Where to run it6 hosts, cheapest first
Claude Sonnet 54 hosts
HostInOutContextUptime
  • Amazon Bedrock$2.00 in·$10.00 out·1M·100% up
  • Azure AI Foundry$2.00 in·$10.00 out·1M·100% up
  • Anthropic$2.00 in·$10.00 out·1M·100% up
  • Google Vertex AI$2.00 in·$10.00 out·1M·100% up
Gemini 3.5 Flash2 hosts
HostInOutContextUptime
  • Google Vertex AI$0.75 in·$4.50 out·1M–not listed
  • Google AI Studio$1.50 in·$9.00 out·1M·100% up

Per million tokens. Prices and uptime via OpenRouter, checked 12 Oct 2026.

Common questions

What is the difference between Claude Sonnet 5 and Gemini 3.5 Flash?

Claude Sonnet 5 is developed by Anthropic while Gemini 3.5 Flash is developed by Google AI. Claude Sonnet 5 has a 1.0M token context window vs Gemini 3.5 Flash's 1.0M. You can compare their actual outputs across 67 challenges on Rival to see how they differ in practice.

Which is better, Claude Sonnet 5 or Gemini 3.5 Flash?

It depends on your use case. Claude Sonnet 5 and Gemini 3.5 Flash each have strengths in different areas. Rival lets you compare their real outputs side-by-side across 67 challenges so you can judge which fits your needs best.

How much does Claude Sonnet 5 cost compared to Gemini 3.5 Flash?

Claude Sonnet 5 costs $2/M input tokens and Gemini 3.5 Flash costs $1.5/M input tokens. Gemini 3.5 Flash is $0.50/M cheaper per input. Check their side-by-side outputs on Rival to see if the price difference is justified by quality.

How can I compare Claude Sonnet 5 and Gemini 3.5 Flash on Rival?

This page shows a side-by-side comparison of Claude Sonnet 5 and Gemini 3.5 Flash 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.

More comparisons

Against the newest arrivals

  • Claude Sonnet 5 vs Muse WonderLanded Oct 2026
  • Gemini 3.5 Flash vs GPT-4 TurboLanded Oct 2026
  • Claude Sonnet 5 vs Step 5 PreviewLanded Oct 2026
  • Gemini 3.5 Flash vs Claude Haiku 5.5Landed Oct 2026
  • Claude Sonnet 5 vs Ling 3.1 FlashLanded Oct 2026
  • Gemini 3.5 Flash vs Mistral Large 4Landed Oct 2026
  • Claude Sonnet 5 vs GPT-6.1 SolLanded Sep 2026
  • Gemini 3.5 Flash vs Claude Sonnet 5.5Landed Sep 2026

Same lab, same size, long tail

  • Claude Sonnet 5 vs Claude Fable 5Version compare
  • Claude Sonnet 5 vs Claude Haiku 5.5Version compare
  • Gemini 3.5 Flash vs Gemini 3.8 FlashSame lab
  • Gemini 3.5 Flash vs Gemini 3.7 FlashSame lab
  • Claude Sonnet 5 vs GPT-6 LunaSame size
  • Claude Sonnet 5 vs GPT-6 Luna ProSame size
  • Claude Sonnet 5 vs GPT-6 SolSame size
  • Claude Sonnet 5 vs GPT-6 Sol ProSame size

Model pages

  • Claude Sonnet 571 outputs, specs and price
  • Gemini 3.5 Flash71 outputs, specs and price
Workshop · 150 skills

Build your own version.

Editable skills and working examples for your agent.

Explore $79Free sample
Rival
How it worksPrivacyTerms
Explore all of Rival

Explore

  • Compare models
  • All models
  • Image comparison
  • Audio comparison
  • Image generation
  • Best AI for...
  • Arena
  • API pricing
  • Challenges

Discover

  • SubjectiveBench
  • Default Index
  • Research
  • Research downloads
  • Rival Kits
  • Find your AI taste
  • UI Glow-Up
  • VoiceLock
  • Cost Cutter
  • Agent skills
  • Benchmarks vs Vibes
  • Brief
  • Jailbreak
  • Model updates
  • Provider status
  • AI creators

Connect

  • Methodology
  • Advertise
  • Partnerships
  • Privacy policy
  • Terms
  • RSS feed