Updated Feb 4, 2026

Our Verdict

Claude Opus 4.6Winner

Mistral Large 3 2512Runner-up

Pick Claude Opus 4.6. In 2 blind votes, Claude Opus 4.6 wins 100% of the time. That's not luck.

Clear winner

Writing DNA

Style Comparison

Similarity

100%

Mistral Large 3 2512 uses 2.4x more emoji

Claude Opus 4.6

Mistral Large 3 2512

53%Vocabulary50%

35wSentence Length20w

0.68Hedging0.48

5.2Bold11.6

2.9Lists5.7

0.21Emoji0.50

1.34Headings1.18

0.03Transitions0.06

Based on 23 + 24 text responses

vs

Ask them anything yourself

Claude Opus 4.6

Mistral Large 3 2512

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

Claude Opus 4.6 vs Grok 3New provider

Claude Opus 4.6 vs OpenAI o3New provider

Claude Opus 4.6 vs Llama 4 MaverickNew provider

Claude Opus 4.6vsMistral Large 3 2512

48 fights queued

Why Claude Opus 4.6?

2 months newerFeb 2026

Why Mistral Large 3 2512?

Dead even. This one's a coin flip.

Claude Opus 4.6Mistral Large 3 2512

Input price

$5.00/M

—

Output price

$25.00/M

—

Context

1.0M

—

Released

Feb 2026

Dec 2025

Favorites

Claude Opus 4.6Mistral Large 3 2512

Movie

2001: A Space Odyssey

2001: A Space Odyssey

The Shawshank Redemption

The Shawshank Redemption

Album

Kind of Blue

OK Computer

Book

Gödel, Escher, Bach

Gödel, Escher, Bach

Douglas R. Hofstadter

La sombra del viento

La sombra del viento

Carlos Ruiz Zafón

City

Istanbul

Tokyo

Game

The Legend of Zelda: Breath of the Wild

The Legend of Zelda: Breath of the Wild

Adventure, Action4.5

The Legend of Zelda: Breath of the Wild

The Legend of Zelda: Breath of the Wild

Adventure, Action4.5

12 / 48

01

Estimate Complexityreasoning

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

Estimating the total number of floating-point operations (FLOPs) required to train GPT-3 involves breaking down the training process into its fundamental components: forward pass, backward pass, and parameter updates. Here's a step-by-step breakdown:

1. Key Parameters of GPT-3

GPT-3 has the following key specifications (from the paper or known details):

Model size (parameters): ~175 billion (1.75 × 10¹¹) parameters.
Training tokens: ~300 billion tokens (3 × 10¹¹).
Batch size: ~3.2 million tokens per batch (3.2 × 10⁶).
Sequence length: ~2048 tokens.
Number of layers: 96 (for the 175B model).
Hidden dimension: ~12,288 (1.2288 × 10⁴).
Number of attention heads: 96.
Optimizer: Adam (or variant), which requires storing additional state (e.g., momentum and variance).

2. FLOPs per Forward Pass

The forward pass of a transformer model involves:

Embedding lookup: Negligible compared to other operations.
Self-attention: For each layer, the self-attention mechanism computes:
- Query, Key, Value projections: 3 × (sequence_length × hidden_dim × hidden_dim) = 3 × (2048 × 12288 × 12288) FLOPs.
- Attention scores: sequence_length × sequence_length × hidden_dim = 2048 × 2048 × 12288 FLOPs.
- Softmax and weighted sum: ~sequence_length × sequence_length × hidden_dim FLOPs.
- Output projection: sequence_length × hidden_dim × hidden_dim = 2048 × 12288 × 12288 FLOPs.
Feed-forward network (FFN): For each layer, the FFN has two linear layers with an expansion factor of 4:
- First layer: sequence_length × hidden_dim × (4 × hidden_dim) = 2048 × 12288 × 49152 FLOPs.
- Second layer: sequence_length × (4 × hidden_dim) × hidden_dim = 2048 × 49152 × 12288 FLOPs.
Layer normalization and residual connections: Negligible.

Simplified FLOPs per Layer:

For one layer, the dominant terms are:

Self-attention: ~6 × (sequence_length × hidden_dim²) = 6 × (2048 × 12288²) ≈ 1.8 × 10¹² FLOPs.
FFN: ~8 × (sequence_length × hidden_dim²) = 8 × (2048 × 12288²) ≈ 2.4 × 10¹² FLOPs.
Total per layer: ~4.2 × 10¹² FLOPs.

For 96 layers: 96 × 4.2 × 10¹² ≈ 4.0 × 10¹⁴ FLOPs per forward pass.

FLOPs per Token:

Since the sequence length is 2048, the FLOPs per token is: 4.0 × 10¹⁴ / 2048 ≈ 2.0 × 10¹¹ FLOPs per token.

3. FLOPs per Backward Pass

The backward pass requires roughly 2× the FLOPs of the forward pass because:

Gradients must be computed for all operations (same as forward).
Additional operations for weight updates (e.g., gradient accumulation).

Thus, backward pass FLOPs ≈ 2 × forward pass FLOPs ≈ 8.0 × 10¹⁴ FLOPs per batch.

4. Total FLOPs per Batch

Forward pass: 4.0 × 10¹⁴ FLOPs.
Backward pass: 8.0 × 10¹⁴ FLOPs.
Total per batch: 1.2 × 10¹⁵ FLOPs.

5. Total Training FLOPs

Number of batches: Total tokens / tokens per batch = 3 × 10¹¹ / 3.2 × 10⁶ ≈ 9.4 × 10⁴ batches.
Total FLOPs: 9.4 × 10⁴ × 1.2 × 10¹⁵ ≈ 1.1 × 10²⁰ FLOPs.

However, this is an underestimate because:

Optimizer overhead: Adam requires storing and updating momentum and variance, which adds ~2× more FLOPs (since each parameter update involves 2 additional operations).
- Total FLOPs with Adam: ~3 × forward + backward FLOPs ≈ 3 × 1.2 × 10¹⁵ = 3.6 × 10¹⁵ per batch.
- Total training FLOPs: 9.4 × 10⁴ × 3.6 × 10¹⁵ ≈ 3.4 × 10²⁰ FLOPs.
Mixed precision training: Often uses FP16/FP32, which may reduce FLOPs slightly, but we'll ignore this for simplicity.

6. Refining the Estimate

A more precise formula for transformer FLOPs (from Kaplan et al., 2020) is: [ \text{Total FLOPs} = 6 \times N \times D ] where:

( N ) = number of parameters (~1.75 × 10¹¹).
( D ) = number of training tokens (~3 × 10¹¹).

This gives: 6 × 1.75 × 10¹¹ × 3 × 10¹¹ ≈ 3.15 × 10²³ FLOPs.

This is much larger than our previous estimate because:

The formula accounts for the fact that each parameter is updated for every token (not just per batch).
It includes the full backward pass and optimizer overhead.

However, this seems too high. A better approximation is: [ \text{Total FLOPs} \approx 2 \times N \times D \times \text{seq_len} ] But this is also not precise.

Empirical Scaling Laws:

From Chinchilla scaling laws, the FLOPs for training a transformer can be approximated as: [ \text{FLOPs} \approx 6 \times N \times D ] For GPT-3: 6 × 1.75 × 10¹¹ × 3 × 10¹¹ ≈ 3.15 × 10²³ FLOPs.

But this is likely an overestimate. The actual number is closer to 3 × 10²³ FLOPs (as reported in the GPT-3 paper).

7. Final Estimate

Combining all the above, the most reasonable estimate is: [ \text{Total training FLOPs for GPT-3} \approx 3.14 \times 10^{23} \text{ FLOPs} ] This aligns with:

The 6 × N × D rule of thumb.
Reported estimates for GPT-3 (e.g., ~3.14 × 10²³ FLOPs in the paper).

Summary of Steps:

Identify model parameters (N = 1.75 × 10¹¹).
Identify training tokens (D = 3 × 10¹¹).
Use the empirical scaling law: FLOPs ≈ 6 × N × D.
Compute: 6 × 1.75 × 10¹¹ × 3 × 10¹¹ ≈ 3.15 × 10²³ FLOPs.

This is the most reliable estimate without looking up the exact number. The actual number may vary slightly due to implementation details (e.g., mixed precision, gradient checkpointing).

Try this prompt

Vote

2 / 48

02

AI Board Game Logicreasoning

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026svg

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025svg

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

Try this prompt

Vote

3 / 48

03

Logic Puzzlereasoning

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

Alright, let's tackle this problem step by step. The question is:

"Sally (a girl) has 3 brothers. Each brother has 2 sisters. How many sisters does Sally have?"

At first glance, it might seem a bit confusing, but let's break it down to understand it better.

Understanding the Problem

We have a family with several children, including Sally and her brothers and sisters. Here's what we know:

Sally is a girl.
Sally has 3 brothers.
Each brother has 2 sisters.

We need to determine how many sisters Sally has.

Identifying the Siblings

First, let's list out the siblings based on the information given.

Sally: She is one girl in the family.
Brothers: Sally has 3 brothers. So, there are 3 boys in the family.

Now, the next piece of information is that "each brother has 2 sisters."

This means that if you pick any one of the 3 brothers, that brother has 2 sisters.

Determining the Sisters

Here's where it might get a bit tricky. We know Sally is one sister, but are there more?

Let's think about what "each brother has 2 sisters" implies.

If a brother has 2 sisters, those sisters must include Sally and possibly another girl.

But wait—Sally is already one sister. So, if a brother has 2 sisters, and one is Sally, who is the other one?

This suggests that there must be at least one more sister besides Sally to make the count of sisters equal to 2 for each brother.

Let's test this:

Suppose there is only Sally as the sister. Then each brother would have only 1 sister (Sally), but the problem states each has 2 sisters. So, this can't be the case.
Therefore, there must be another sister besides Sally.

So, the sisters are:

Sally
Another girl (let's call her, say, Emily)

Now, let's check if this satisfies the condition that each brother has 2 sisters.

Each brother's sisters would be:

Sally
Emily

That's 2 sisters, which matches the given information.

Verifying the Number of Sisters

Now, the question is asking: How many sisters does Sally have?

We've established that the sisters in the family are:

Sally
Emily

But the question is about Sally's sisters, not including herself.

So, Sally's sisters are:

Emily

That's 1 sister.

Wait a minute, but earlier we thought there are 2 sisters in total (Sally and Emily), but Sally's sisters would be just Emily, making it 1 sister.

But let's double-check the initial interpretation to ensure we didn't make a mistake.

Re-evaluating the Interpretation

The confusion might arise from how we're counting the sisters.

Total sisters in the family:

Sally
Emily

Total brothers: 3

Now, "each brother has 2 sisters." Each brother's sisters are Sally and Emily, which is 2. This checks out.

Now, how many sisters does Sally have?

Sally's sisters are the other girls in the family besides herself, which is Emily. So, Sally has 1 sister.

But let me think if there could be more sisters.

Suppose there are 3 sisters: Sally, Emily, and another girl (let's say, Anna).

Then each brother would have 3 sisters (Sally, Emily, Anna), but the problem says each has 2 sisters. So, this can't be.

Therefore, the only possibility is that there are 2 sisters in total: Sally and Emily.

Thus, Sally has 1 sister.

Potential Misinterpretation

Initially, one might think that since each brother has 2 sisters, and there are 3 brothers, there are 6 sister instances (3 brothers × 2 sisters each), leading to more sisters. But this would be incorrect because all brothers share the same sisters; the sisters aren't unique to each brother.

For example, if there are 2 sisters, each of the 3 brothers has those 2 sisters, totaling 6 sister instances (3 brothers × 2 sisters), but the actual number of sisters is just 2.

Final Confirmation

Let's confirm with the numbers:

Sisters: Sally and Emily (total 2 girls)
Brothers: 3 boys

Each brother's sisters: Sally and Emily (2 sisters) — matches the given information.

Sally's sisters: Emily (1 sister)

Therefore, Sally has 1 sister.

Conclusion

After carefully analyzing the problem and considering the relationships, the correct answer is:

Sally has 1 sister.

Try this prompt

Vote

4 / 48

04

Stochastic Consistency Testanalysis

Tests an AI's randomness and creativity

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

Here are five jokes for you:

Why don’t skeletons fight each other? They don’t have the guts.
Why did the scarecrow win an award? Because he was outstanding in his field!
What do you call a fake noodle? An impasta.
Why can’t you trust an atom? Because they make up everything!
Why did the bicycle fall over? Because it was two-tired!

Uniqueness Check:

All five jokes are unique—they’re distinct in setup, punchline, and theme. No repeats here!

Want more? I can generate even weirder ones. 😄

Try this prompt

Vote

5 / 48

05

SVG Layout Challengeimage generation

Tests an AI's ability to generate vector graphics

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026svg

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025svg

Try this prompt

Vote

6 / 48

06

Xbox Controller SVG Artimage generation

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026svg

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025svg

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

Try this prompt

Vote

Sponsored

7 / 48

07

Generate a Stand-Up Routineconversation

Tests an AI's humor and creative writing ability

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

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

Try this prompt

Vote

8 / 48

08

Realistic AI Interviewconversation

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

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

Try this prompt

Vote

9 / 48

09

Satirical Fake News Headlineconversation

Tests an AI's humor and understanding of current events

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

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

Try this prompt

Vote

10 / 48

10

Character Voice Testconversation

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026text

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025text

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

Try this prompt

Vote

11 / 48

11

Minimalist Landing Pageweb design

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026website

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025website

Try this prompt

Vote

12 / 48

12

Pokémon Battle UI Recreationweb design

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

Claude Opus 4.6

Claude Opus 4.6

Anthropic · Feb 2026website

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

Try this prompt

Mistral Large 3 2512

Mistral Large 3 2512

Mistral · Dec 2025website

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

Try this prompt

Vote

This matchup has more rounds

36+ 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