Updated May 6, 2025

Our Verdict

Gemini 2.5 Pro (I/O Edition)Winner

GPT-4o (Omni)Runner-up

Pick Gemini 2.5 Pro (I/O Edition). In 15 blind votes, Gemini 2.5 Pro (I/O Edition) wins 100% of the time. That's not luck.

Gemini 2.5 Pro (I/O Edition) particularly excels in Image Generation, Web Design.

Clear winner

Writing DNA

Style Comparison

Similarity

97%

GPT-4o (Omni) uses 2.5x more emoji

GPT-4o (Omni)

Gemini 2.5 Pro (I/O Edition)

54%Vocabulary58%

18wSentence Length16w

0.72Hedging0.38

7.3Bold5.8

5.6Lists5.0

0.03Emoji0.00

1.40Headings0.63

0.26Transitions0.01

Based on 24 + 16 text responses

vs

Ask them anything yourself

GPT-4o (Omni)

Gemini 2.5 Pro (I/O Edition)

279 AI models invented the same fake scientist.

We read every word. 250 models. 2.14 million words. This is what we found.

AI Hallucination Index 2026

Free preview13 of 58 slides

Download the free preview or get all 58 slides for $49

FAQ

Common questions

Keep going

GPT-4o (Omni) vs Llama 4 MaverickNew provider

Gemini 2.5 Pro (I/O Edition) vs GPT-5Cross-provider

GPT-4o (Omni) vs Grok 3New provider

GPT-4o (Omni)vsGemini 2.5 Pro (I/O Edition)

34 fights queued

Why GPT-4o (Omni)?

Dead even. This one's a coin flip.

Why Gemini 2.5 Pro (I/O Edition)?

12 months newerMay 2025

GPT-4o (Omni)Gemini 2.5 Pro (I/O Edition)

Input price

$2.50/M

$1.25/M

Output price

$10.00/M

$10.00/M

Context

128K

—

Released

May 2024

May 2025

Favorites

GPT-4o (Omni)Gemini 2.5 Pro (I/O Edition)

Movie

Inception

Inception

2001

2001

Album

Dark Side Of The Moon

Dark Side Of The Moon

The Dark

Book

To Kill a Mockingbird

To Kill a Mockingbird

The Hitch Hiker's Guide to the Galaxy

The Hitch Hiker's Guide to the Galaxy

City

Kyoto

Kyoto

Game

The Legend of Zelda: Breath of the Wild

The Legend of Zelda: Breath of the Wild

Adventure, Action4.5

CHRONO

CHRONO

12 / 34

01

Estimate Complexityreasoning

Tests an AI's ability to make educated estimates based on technical knowledge

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

Okay, let's estimate the FLOPs for training GPT-3. I'll break this down.

Key Formula: The number of FLOPs for training a transformer-based model can be roughly estimated as: FLOPs ≈ 6 * N * D Where:

N is the number of parameters in the model.
D is the total number of tokens processed during training (dataset size in tokens * number of epochs).
The factor of 6 comes from:
- 2 * N FLOPs per token for the forward pass (each parameter is involved in roughly one multiply-accumulate, and a MAC is 2 FLOPs).
- 4 * N FLOPs per token for the backward pass (gradient computation is roughly twice as expensive as the forward pass).

Assumptions for GPT-3 (Davinci, the largest version):

Number of Parameters (N):
- GPT-3 (the largest variant, Davinci) is famously stated to have 175 billion parameters.
- So, N = 175 * 10^9.
Number of Tokens Processed (D):
- This is the trickier part to recall without searching. I remember it was trained on a massive amount of data, a significant portion of Common Crawl, books, Wikipedia, etc.
- Large Language Models are typically trained for roughly one epoch over their massive datasets, or even less if datasets are mixed and weighted. The key figure is the total number of tokens seen by the model.
- I seem to recall a figure in the range of 300 billion to 500 billion tokens for GPT-3's training. The number often cited in discussions around the GPT-3 paper and subsequent scaling law papers (like Chinchilla) is around 300 billion tokens. Let's use that as a reasonable estimate for D.
- So, D = 300 * 10^9.

Calculation:

Now, let's plug these into the formula: FLOPs ≈ 6 * N * D FLOPs ≈ 6 * (175 * 10^9) * (300 * 10^9)

Let's do the numerical part: 6 * 175 = 1050 1050 * 300 = 315,000

Now, the powers of 10: 10^9 * 10^9 = 10^18

So, combining these: FLOPs ≈ 315,000 * 10^18 FLOPs ≈ 3.15 * 10^5 * 10^18 FLOPs ≈ 3.15 * 10^(5+18) FLOPs ≈ 3.15 * 10^23

Expressing the Result:

TeraFLOPs (TFLOPs) = 10^12
PetaFLOPs (PFLOPs) = 10^15
ExaFLOPs (EFLOPs) = 10^18
ZettaFLOPs (ZFLOPs) = 10^21
YottaFLOPs (YFLOPs) = 10^24

So, 3.15 * 10^23 FLOPs can be written as:

315 * 10^21 FLOPs = 315 ZettaFLOPs
Or, as 0.315 * 10^24 FLOPs = 0.315 YottaFLOPs

The number is often quoted in PetaFLOP/s-days or similar. Let's convert 3.15 * 10^23 FLOPs to PetaFLOP-seconds: 3.15 * 10^23 / 10^15 = 3.15 * 10^8 PetaFLOP-seconds.

Number of seconds in a day = 24 hours/day * 60 minutes/hour * 60 seconds/minute = 86,400 seconds/day. So, 3.15 * 10^8 PetaFLOP-seconds / 86,400 seconds/day ≈ 3.15 * 10^8 / (8.64 * 10^4) PetaFLOP-days ≈ (3.15 / 8.64) * 10^(8-4) PetaFLOP-days ≈ 0.3646 * 10^4 PetaFLOP-days ≈ 3646 PetaFLOP/s-days (if executed on a machine capable of 1 PFLOP/s constantly for that many days, or scaled accordingly).

Reasoning Summary:

Identified the core formula: FLOPs ≈ 6 * N * D for transformer training.
Recalled model parameters (N): GPT-3 Davinci has 175 billion parameters.
Estimated training tokens (D): Based on common knowledge about LLM training and GPT-3 specifics, estimated ~300 billion tokens processed. This assumes roughly one pass over the (weighted) training data.
Performed the multiplication: 6 * (175 * 10^9) * (300 * 10^9) = 3.15 * 10^23 FLOPs.
Interpreted the result: This is approximately 315 ZettaFLOPs. This also aligns with the often-cited figure for GPT-3 training compute, which is in the range of 3000-4000 PetaFLOP/s-days (my calculation of ~3646 PetaFLOP/s-days is right in this ballpark).

Caveats:

The 6ND is an approximation. The exact number can vary slightly based on the specific transformer architecture details (e.g., use of sparse attention, specific normalization layers), optimizer overhead, and whether techniques like activation checkpointing are used (which can increase FLOPs for recomputation but save memory).
Mixed-precision training (e.g., bfloat16 or fp16) doesn't change the number of FLOPs, but it changes the type of FLOPs and allows for faster computation on hardware that supports it.
The number of training tokens D is an estimate. The GPT-3 paper describes a complex data mixture and weighting schedule, so "300 billion tokens" is an effective number of tokens processed.

Therefore, my estimate is approximately 3.15 x 10^23 FLOPs or 315 ZettaFLOPs.

Try this prompt

Vote

2 / 34

02

Logic Puzzlereasoning

Tests an AI's ability to solve a simple but potentially confusing logic puzzle

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

Let's break this down:

Sally is a girl.
Sally has 3 brothers.
These 3 brothers are all part of the same family.
Each brother shares the same sisters.
If one brother has 2 sisters, those 2 sisters are Sally and one other girl.

So, Sally has 1 sister.

Try this prompt

Vote

3 / 34

03

Math Misconception Testreasoning

Tests an AI's understanding of number representation

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

No, 9.11 is not greater than 9.9.

Here's why:

Compare the whole number part: Both numbers have 9 as the whole number p art, so they are equal so fa r. 2. Compare the tenths place (the first digit afte r the decimal):
- 9.11 has 1 in the tenths place.
- 9.9 has 9 in the tenths place.
Since 1 is less than 9, 9.11 is less than 9.9.

You can also think of it as 9.9 being the same as 9.90. Comparing 9.11 and 9.90, it's easier to see that 11 (hundredths) is less than 90 (hundredths).

Try this prompt

Vote

4 / 34

04

Stochastic Consistency Testanalysis

Tests an AI's randomness and creativity

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

Here are 5 jokes:

Why don't scientists trust atoms? Because they make up everything!
What do you call a fish with no eyes? Fsh!
I told my wife she was drawing her eyebrows too high. She seemed surprised.
Why did the scarecrow win an award? Because he was outstanding in his field!
What do you call a lazy kangaroo? Pouch potato!

All 5 of these jokes are unique.

Try this prompt

Vote

5 / 34

05

SVG Layout Challengeimage generation

Tests an AI's ability to generate vector graphics

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024svg

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025svg

Try this prompt

Vote

6 / 34

06

Xbox Controller SVG Artimage generation

Tests an AI's ability to create detailed SVG illustrations of gaming hardware

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024svg

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025svg

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Vote

Sponsored

7 / 34

07

Generate a Stand-Up Routineconversation

Tests an AI's humor and creative writing ability

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Vote

8 / 34

08

Realistic AI Interviewconversation

Tests an AI's ability to simulate personalities and predict future trends

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Vote

9 / 34

09

Satirical Fake News Headlineconversation

Tests an AI's humor and understanding of current events

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025text

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Vote

10 / 34

10

Minimalist Landing Pageweb design

Tests an AI's ability to generate a complete, working landing page

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024website

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025website

Try this prompt

Vote

11 / 34

11

Pokémon Battle UI Recreationweb design

Recreate an interactive, nostalgic Pokémon battle UI in a single HTML file.

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024website

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025website

Nothing here. The model returned empty. We stared at it for a while.

Try this prompt

Vote

12 / 34

12

Mario Level UI Recreationweb design

Recreate an interactive, classic Mario level in a single HTML file.

GPT-4o (Omni)

GPT-4o (Omni)

Openai · May 2024website

Try this prompt

Gemini 2.5 Pro (I/O Edition)

Gemini 2.5 Pro (I/O Edition)

Google · May 2025website

Try this prompt

Vote

This matchup has more rounds

22+ 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