What is the difference between Claude Sonnet 4.5 and GPT-5?

Claude Sonnet 4.5 is developed by Anthropic while GPT-5 is developed by OpenAI. Claude Sonnet 4.5 has a 200K token context window vs GPT-5's 400K. in 53 community votes on RIVAL, Claude Sonnet 4.5 wins 64% of head-to-head matchups. These results are based on blind head-to-head voting across 37 challenges.

Which is better, Claude Sonnet 4.5 or GPT-5?

Based on 53 community votes on RIVAL, Claude Sonnet 4.5 wins 64% of head-to-head matchups against GPT-5. Claude Sonnet 4.5 is strongest in Web Design, Reasoning. However, GPT-5 leads in Image Generation, Conversation.

How much does Claude Sonnet 4.5 cost compared to GPT-5?

Claude Sonnet 4.5 costs $3/M input tokens and GPT-5 costs $1.25/M input tokens. GPT-5 is $1.75/M cheaper per input. The more expensive model wins 64% of duels, so the premium may be justified by quality.

How are Claude Sonnet 4.5 vs GPT-5 votes collected on RIVAL?

RIVAL presents both models' outputs side-by-side in blind duels — voters see the responses but don't know which model produced each one. This eliminates brand bias. 53 votes have been collected for this pair across 37 challenges. All vote data is part of RIVAL's open dataset.

Rival

Claude Sonnet 4.5vsGPT-5

37 challenges

Why Claude Sonnet 4.5?

Leads 1 of 1 benchmarks

Stronger on SWE-bench Verified77.2% vs 74.9%

2 months newerSep 2025

Why GPT-5?

1.6x cheaper overall$1.25/M in · $10.00/M out

2x more context400K

Favorites

Claude Sonnet 4.5GPT-5

Movie

2001: A Space Odyssey

19688.1

—

Album

Kind of Blue

Miles Davis

—

Book

Die Verwandlung

Franz Kafka

—

City

Amsterdam

Netherlands

—

Game

Portal 2

Shooter, Puzzle4.6

—

12 / 37

Estimate Complexityreasoning

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

Claude Sonnet 4.5

Anthropic · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

Here’s a back‑of‑the‑envelope estimate using standard LM training-compute heuristics.

Assumptions

Model size (N): GPT-3 has about 175 billion parameters (N ≈ 1.75 × 10^11).
Training tokens (T): GPT-3 was trained on roughly 300 billion tokens (T ≈ 3.0 × 10^11).
FLOPs per token per parameter: A common rule of thumb for transformer LM training is FLOPs ≈ 6 × N × T The factor ~6 comes from counting forward and backward passes for the parameter-heavy matmuls (roughly 2N for forward + 4N for backward). It ignores small terms from layernorms, softmax, embeddings, and optimizer updates (which we’ll account for with a small overhead factor later).

Core calculation

6 × N × T = 6 × (1.75 × 10^11) × (3.0 × 10^11)
First multiply N × T: 1.75 × 3.0 = 5.25; powers: 10^11 × 10^11 = 10^22 So N × T = 5.25 × 10^22
Now multiply by 6: 6 × 5.25 × 10^22 = 31.5 × 10^22 = 3.15 × 10^23 FLOPs

Overheads and uncertainty

Non-matmul ops (attention softmax, masking, residuals, layernorms), embeddings, and optimizer update steps add some overhead. A 10–20% uplift is a reasonable allowance.
Applying +10–20%: 3.15 × 10^23 × (1.1 to 1.2) ≈ (3.5 ± 0.3) × 10^23 FLOPs.

Answer

Point estimate: ~3.15 × 10^23 FLOPs
With a modest overhead allowance: roughly (3.3–3.8) × 10^23 FLOPs

This is a coarse estimate; the exact number depends on details like sequence length, activation checkpointing, implementation efficiency, and the precise token count.

Vote

2 / 37

AI Board Game Logicreasoning

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

Claude Sonnet 4.5

Anthropic · Sep 2025svg

No content available for this output.

GPT-5

Openai · Aug 2025svg

No content available for this output.

Vote

3 / 37

Logic Puzzlereasoning

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

Claude Sonnet 4.5

Anthropic · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

Reason: If each of the 3 brothers has 2 sisters, there are exactly 2 girls in the family. One is Sally, so she has 1 sister.

Vote

4 / 37

Math Misconception Testreasoning

Tests an AI's understanding of number representation

Claude Sonnet 4.5

Anthropic · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No. 9.11 is less than 9.9. Think of 9.9 as 9.90; comparing 9.11 to 9.90 shows 11 hundredths < 90 hundredths.

Vote

5 / 37

Stochastic Consistency Testanalysis

Tests an AI's randomness and creativity

Claude Sonnet 4.5

Anthropic · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

I told my suitcases we’re not going on vacation this year. Now I’m dealing with emotional baggage.
Parallel lines have so much in common. It’s a shame they’ll never meet.
Why do programmers prefer dark mode? Because light attracts bugs.
I ordered a chicken and an egg online. I’ll let you know which comes first.
I asked the librarian if the library had any books on paranoia. She whispered, “They’re right behind you.”

Unique jokes: 5

Vote

6 / 37

SVG Layout Challengeimage generation

Tests an AI's ability to generate vector graphics

Claude Sonnet 4.5

Anthropic · Sep 2025svg

No content available for this output.

GPT-5

Openai · Aug 2025svg

Vote

7 / 37

Xbox Controller SVG Artimage generation

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

Claude Sonnet 4.5

Anthropic · Sep 2025svg

No content available for this output.

GPT-5

Openai · Aug 2025svg

No content available for this output.

Vote

8 / 37

Generate a Stand-Up Routineconversation

Tests an AI's humor and creative writing ability

Claude Sonnet 4.5

Anthropic · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

9 / 37

Character Voice Testconversation

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

Claude Sonnet 4.5

Anthropic · Sep 2025text

No content available for this output.

GPT-5

Openai · Aug 2025text

No content available for this output.

Vote

10 / 37

Pokémon Battle UI Recreationweb design

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

Claude Sonnet 4.5

Anthropic · Sep 2025website

No content available for this output.

GPT-5

Openai · Aug 2025website

No content available for this output.

Vote

11 / 37

Mario Level UI Recreationweb design

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

Claude Sonnet 4.5

Anthropic · Sep 2025website

GPT-5

Openai · Aug 2025website

Vote

12 / 37

Linear App Cloneweb design

Tests an AI's ability to replicate an existing UI with Tailwind CSS

Claude Sonnet 4.5

Anthropic · Sep 2025website

No content available for this output.

GPT-5

Openai · Aug 2025website

No content available for this output.

Vote

12 of 37

FAQ

Continue exploring

Claude Sonnet 4.5 vs Llama 4 MaverickNew provider

Claude Sonnet 4.5 vs MiniMax M2.5New provider

Claude Sonnet 4.5 vs Pony AlphaNew provider

Here’s a back‑of‑the‑envelope estimate using standard LM training-compute heuristics.

Assumptions

Model size (N): GPT-3 has about 175 billion parameters (N ≈ 1.75 × 10^11).
Training tokens (T): GPT-3 was trained on roughly 300 billion tokens (T ≈ 3.0 × 10^11).
FLOPs per token per parameter: A common rule of thumb for transformer LM training is FLOPs ≈ 6 × N × T The factor ~6 comes from counting forward and backward passes for the parameter-heavy matmuls (roughly 2N for forward + 4N for backward). It ignores small terms from layernorms, softmax, embeddings, and optimizer updates (which we’ll account for with a small overhead factor later).

Core calculation

6 × N × T = 6 × (1.75 × 10^11) × (3.0 × 10^11)
First multiply N × T: 1.75 × 3.0 = 5.25; powers: 10^11 × 10^11 = 10^22 So N × T = 5.25 × 10^22
Now multiply by 6: 6 × 5.25 × 10^22 = 31.5 × 10^22 = 3.15 × 10^23 FLOPs

Overheads and uncertainty

Non-matmul ops (attention softmax, masking, residuals, layernorms), embeddings, and optimizer update steps add some overhead. A 10–20% uplift is a reasonable allowance.
Applying +10–20%: 3.15 × 10^23 × (1.1 to 1.2) ≈ (3.5 ± 0.3) × 10^23 FLOPs.

Answer

Point estimate: ~3.15 × 10^23 FLOPs
With a modest overhead allowance: roughly (3.3–3.8) × 10^23 FLOPs

This is a coarse estimate; the exact number depends on details like sequence length, activation checkpointing, implementation efficiency, and the precise token count.