Updated Aug 7, 2025

Our Verdict

DeepSeek R1Winner

GPT-5 MiniRunner-up

DeepSeek R1 has the edge overall. In 11 blind votes, DeepSeek R1 wins 57% of the time.

Pick DeepSeek R1 for Web Design. Pick GPT-5 Mini for Image Generation.

Slight edge

Writing DNA

Style Comparison

Similarity

85%

DeepSeek R1 uses 800.0x more bold

DeepSeek R1

GPT-5 Mini

62%Vocabulary62%

15wSentence Length21w

0.52Hedging0.16

8.0Bold0.0

5.7Lists3.4

0.04Emoji0.00

0.62Headings0.00

0.28Transitions0.05

Based on 22 + 15 text responses

vs

Ask them anything yourself

DeepSeek R1

GPT-5 Mini

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

DeepSeek R1 vs MiniMax M2.5New provider

DeepSeek R1 vs Aurora AlphaNew provider

GPT-5 Mini vs Claude Sonnet 4New provider

DeepSeek R1vsGPT-5 Mini

46 fights queued

Why DeepSeek R1?

Dead even. This one's a coin flip.

Why GPT-5 Mini?

1.2x cheaper overall$0.25/M in · $2.00/M out

6 months newerAug 2025

DeepSeek R1GPT-5 Mini

Input price

$0.55/M

$0.25/M

Output price

$2.19/M

$2.00/M

Context

128K

—

Released

Feb 2025

Aug 2025

Favorites

DeepSeek R1GPT-5 Mini

Movie

The Godfather

The Godfather

—

Album

The Dark Side of the Moon

The Dark Side of the Moon

—

Book

Pride and Prejudice

Pride and Prejudice

—

City

Kyoto

—

Game

The Legend of Zelda: Ocarina of Time

The Legend of Zelda: Ocarina of Time

Adventure, Action4.4

—

12 / 46

01

Estimate Complexityreasoning

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

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

Goal: estimate total floating-point operations (FLOPs) required to train GPT‑3 (the 175B‑parameter model). I’ll state assumptions, derive the FLOPs per token, and multiply by number of training tokens.

Assumptions (reasonably standard for GPT‑3):

Model size: N ≈ 175 billion parameters (175 × 10^9).
Transformer shape consistent with that size: L = 96 layers, model width d = 12,288 (these give ≈175B params; I’ll use d and L only to justify a FLOP count check).
Context / sequence length used in training: S = 2048.
Number of training tokens: T ≈ 300 billion tokens (3 × 10^11). (This is an approximate dataset scale used in large‑language training; changing this linearly scales compute.)
Backpropagation / training overhead: backward pass + weight gradient computation ≈ 2× the forward cost, so total training cost ≈ 3× forward cost (forward + backward + a small extra for optimizer).

Step 1 — Rough forward cost per token (per layer) Consider one transformer layer acting on a sequence of S tokens, width d:

Q/K/V linear projections: 3 × (S × d) × d matmuls → about 6·S·d^2 FLOPs (counting multiply+add as 2 FLOPs).
Output projection from attention: (S × d) × d → 2·S·d^2 FLOPs. → Total for projections ≈ 8·S·d^2.
Attention core (Q·K^T and attention-weighted sum): two big matmuls of shapes (S × d) × (d × S) and (S × S) × (S × d) → about 4·S^2·d FLOPs.
Feed‑forward (two linear layers d → 4d → d): 2·S·d·(4d) + 2·S·(4d)·d = 16·S·d^2 FLOPs.

So per layer forward FLOPs for a sequence of length S: F_forward_layer ≈ 8·S·d^2 + 4·S^2·d + 16·S·d^2 = 24·S·d^2 + 4·S^2·d.

Divide by S to get forward FLOPs per token per layer: f_layer_token ≈ 24·d^2 + 4·S·d.

Step 2 — Plug in numbers (d = 12,288, S = 2048, L = 96)

d^2 = 12,288^2 = 150,994,944 ≈ 1.50995 × 10^8.
24·d^2 ≈ 24 × 1.50995e8 ≈ 3.6239 × 10^9.
4·S·d = 4 × 2048 × 12,288 = 100,663,296 ≈ 1.0066 × 10^8.
So f_layer_token ≈ 3.6239e9 + 1.0066e8 ≈ 3.7246 × 10^9 FLOPs per token per layer.

Multiply by L = 96 layers:

forward FLOPs per token for whole model ≈ 96 × 3.7246e9 ≈ 3.5756 × 10^11 ≈ 3.58 × 10^11 FLOPs.

Step 3 — Include backward pass / training overhead

Training (forward + backward + grads) ≈ 3 × forward FLOPs (typical rule of thumb).
So training FLOPs per token ≈ 3 × 3.58 × 10^11 ≈ 1.07 × 10^12 FLOPs per token.

Step 4 — Multiply by total tokens

T = 3 × 10^11 tokens → Total FLOPs ≈ 1.07 × 10^12 × 3 × 10^11 ≈ 3.2 × 10^23 FLOPs.

Shortcut / simple rule A common simple approximation comes out of the algebra above: training FLOPs ≈ 6 × N_params × N_tokens (because the forward FLOPs per token are ≈ 2×N_params, and training is about 3× forward → ~6×N_params). Using that:

6 × 175×10^9 × 300×10^9 = 6 × 175 × 300 × 10^18 = 315,000 × 10^18 = 3.15 × 10^23 FLOPs, which is the same ballpark.

Final estimate

≈ 3 × 10^23 FLOPs (more precisely ≈ 3.1–3.3 × 10^23 FLOPs) to train GPT‑3 (175B) on ≈300B tokens with S=2048.

Caveats and sensitivity

If the dataset contained more tokens (e.g., 400–500B) the total scales linearly (e.g., 500B tokens → ≈5.2 × 10^23 FLOPs).
If you use a smaller context length during training, the S-dependent attention term (4·S^2·d per layer) shrinks, but for large d the 24·d^2 term dominates so S matters less (unless S is much larger).
Optimizer choice, precision (fp16 vs fp32), and implementation efficiencies (fused kernels, reduced-memory backprop techniques) can change actual FLOP counts and wall‑clock compute by tens of percent.
This is an estimate of raw FLOPs; real training time also depends on parallelization, communication overhead, and hardware efficiency.

Summary: roughly 3 × 10^23 floating-point operations to train GPT‑3 (175B) under the stated assumptions.

Try this prompt

Vote

2 / 46

02

AI Board Game Logicreasoning

Tests an AI's ability to understand game rules and strategy

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025svg

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025svg

Try this prompt

Vote

3 / 46

03

Logic Puzzlereasoning

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

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

1

Explanation: The three brothers share the same two sisters (Sally and one other girl). Counting sisters for Sally excludes herself, so she has one sister.

Try this prompt

Vote

4 / 46

04

Math Misconception Testreasoning

Tests an AI's understanding of number representation

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

No — as usual decimal numbers, 9.11 is less than 9.9.

Reason: 9.11 = 9.11 and 9.9 = 9.90, and 0.11 < 0.90, so 9.11 < 9.9.

(If you meant software versioning, 9.11 is typically considered greater than 9.9 because 11 > 9.)

Try this prompt

Vote

5 / 46

05

Stochastic Consistency Testanalysis

Tests an AI's randomness and creativity

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

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

Try this prompt

Vote

6 / 46

06

SVG Layout Challengeimage generation

Tests an AI's ability to generate vector graphics

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025svg

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025svg

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

Try this prompt

Vote

Sponsored

7 / 46

07

Xbox Controller SVG Artimage generation

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

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025svg

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025svg

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

Try this prompt

Vote

8 / 46

08

Generate a Stand-Up Routineconversation

Tests an AI's humor and creative writing ability

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

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

Try this prompt

Vote

9 / 46

09

Realistic AI Interviewconversation

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

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

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

Try this prompt

Vote

10 / 46

10

Satirical Fake News Headlineconversation

Tests an AI's humor and understanding of current events

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

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

Try this prompt

Vote

11 / 46

11

Character Voice Testconversation

Tests an AI's ability to write in distinct character voices

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025text

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

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025text

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

Try this prompt

Vote

12 / 46

12

Minimalist Landing Pageweb design

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

DeepSeek R1

DeepSeek R1

Deepseek · Feb 2025website

Try this prompt

GPT-5 Mini

GPT-5 Mini

Openai · Aug 2025website

Try this prompt

Vote

This matchup has more rounds

34+ 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