The Known Good Updated 25 Jul 2026
claude-3-5-sonnet-20241022 vs gpt-4.5-preview ·Anthropic vs OpenAI

claude-3-5-sonnet-20241022 vs gpt-4.5-preview — head-to-head comparison

claude-3-5-sonnet-20241022 profile gpt-4.5-preview profile All comparisons
claude-3-5-sonnet-20241022 — Anthropic
Quality
#45 of 61
40based on 3 of 6 evaluations
Known Good Index
Speed
unranked
Output tokens per second
Price
unranked
Blended per 1M tokens
Cache hit price
unranked
per 1M tokens
Context
unranked
tokens
gpt-4.5-preview — OpenAI
Quality
#43 of 61
47based on 4 of 6 evaluations
Known Good Index
Speed
unranked
Output tokens per second
Price
unranked
Blended per 1M tokens
Cache hit price
unranked
per 1M tokens
Context
unranked
tokens
Summary

On the Known Good Index, claude-3-5-sonnet-20241022 scores 40 and gpt-4.5-preview scores 47 — a 7-point lead for gpt-4.5-preview. They share 10 published evaluations, with claude-3-5-sonnet-20241022 ahead on 0 and gpt-4.5-preview ahead on 10.

claude-3-5-sonnet-20241022 is proprietary; gpt-4.5-preview is proprietary.

Generated from tracked data on 25 Jul 2026 · how these comparisons are computed →

At a glance
Metricclaude-3-5-sonnet-20241022gpt-4.5-preview
Known Good Index 40 based on 3 of 6 evaluations 47 based on 4 of 6 evaluations
Blended $/1M
Output tok/s
Context window
Providers 0 0
Jump to section

Quality

Known Good Index and the capability indices, side by side

IndicesComponents
Index scores

Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.

Metricclaude-3-5-sonnet-20241022gpt-4.5-preview
Known Good Index 40 47
Mathematics Index 33 59
Science Index 52 68
Reasoning Index 52 48
Openness Index 5 5
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = claude-3-5-sonnet-20241022 · marker = gpt-4.5-preview · scores ingested from Epoch AI (CC-BY 4.0), LiveBench (Apache-2.0)

GPQA Diamond55 / 69
MATH / AIME8 / 38
LiveBench59 / 69
LiveBench Coding67 / 75
LiveBench Data Analysis55 / 64
LiveBench Instruction Following69 / 72
LiveBench Language54 / 61
LiveBench Mathematics52 / 69
LiveBench Reasoning57 / 71
MATH Level 557 / 79

Right-hand figures read claude-3-5-sonnet-20241022 / gpt-4.5-preview. claude-3-5-sonnet-20241022 leads on 0 of 10, gpt-4.5-preview on 10.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Arena ratings

Elo from blind pairwise votes, ingested from Arena. Higher is better.

Metricclaude-3-5-sonnet-20241022gpt-4.5-preview
Text Arena 1,298
Text Arena (style controlled) 1,373
The Known Good

Price & Cost

Published API pricing by token type

Per token typeCache discount
Published pricing

USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.

Metricclaude-3-5-sonnet-20241022gpt-4.5-preview
Input $/1M
Output $/1M
Blended $/1M
Cache read $/1M
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input accepted and maximum output emitted

Context and output limits

Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.

Metricclaude-3-5-sonnet-20241022gpt-4.5-preview
Context window
Max output tokens
The Known Good

Speed & Latency

Median throughput and time to first token across hosting providers

Output speedTime to first token
Throughput and latency

Medians across all tracked providers, derived from OpenRouter endpoint telemetry.

Metricclaude-3-5-sonnet-20241022gpt-4.5-preview
Output tokens/s
Time to first token
The Known Good

Providers

Which hosts serve each model, and at what price and speed

0 hosts
Full provider benchmarks →

No hosting providers are currently tracked for either model.

The Known Good
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Benchmarks — LiveBench (Apache-2.0)Elo — Arena Full attribution →