Noometry

Model comparison

Claude Sonnet 5.5 vs GPT-5.3 Chat

Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 42.8 on the Noometry Index.

Last verified . 12 shared benchmarks.

Claude Sonnet 5.5 Anthropic

61.9

Rank #10 Confirmed

GPT-5.3 Chat OpenAI

42.8

Rank #109 Confirmed

Summary

  • They share 12 benchmarks with published results for both. Claude Sonnet 5.5 scores higher in 8 categories and GPT-5.3 Chat in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Sonnet 5.5 leads 87.9 to 38.2.
  • Claude Sonnet 5.5 is cheaper at $2 / $10 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Chat.
  • Claude Sonnet 5.5 accepts more context: 1M tokens versus 128K.

Side by side

Claude Sonnet 5.5GPT-5.3 Chat
ProviderAnthropicOpenAI
Noometry Index61.942.8
Released2026-09-282026-03-03
WeightsProprietaryProprietary
Context window1M128K
Max output128K16K
Input $ / M tokens$2$1.75
Output $ / M tokens$10$14
Results tracked3218

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 67.3 (#6), GPT-5.3 Chat: 41.4 (#124)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Coding15131408
FrontierCode52.1%—
CursorBench55.5%—
LMArena WebDev1774—
FrontierSWE61.9%—
SciCode61%—
ALE-Bench1,819—

Agentic & Tool Use Not comparable

Claude Sonnet 5.5: 45.0 (#16), GPT-5.3 Chat: —

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
APEX-Agents75.5%—

Reasoning Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 54.0 (#28), GPT-5.3 Chat: 28.5 (#102)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Hard Prompts14951399
NYT Connections (extended)80.5%—
CritPt31.4%—
Mystery Game Puzzles65%—
Epoch Capabilities Index165.03—

Math Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 87.9 (#6), GPT-5.3 Chat: 38.2 (#142)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Math15101389
FrontierMath (Tiers 1-3)88.8%—
FrontierMath Tier 480.5%—
OTIS Mock AIME 2024-2025100%—
ProofBench100%—
FrontierMath Erdős2.9%—

Knowledge Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 66.0 (#12), GPT-5.3 Chat: 38.8 (#140)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Expert15401397
GPQA Diamond95.6%—
SimpleQA Verified46.5%—

Multimodal Not comparable

Claude Sonnet 5.5: 51.5 (#6), GPT-5.3 Chat: —

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Vision1289—
Furniture Assembly75%—

Multilingual Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 55.3 (#30), GPT-5.3 Chat: 50.3 (#124)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Non-English14521382
LMArena Chinese15221432
LMArena Russian14511400
LMArena French—1397
LMArena German—1384
LMArena Japanese—1352
LMArena Korean—1346
LMArena Spanish—1371

Instruction Following Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 78.3 (#11), GPT-5.3 Chat: 72.8 (#129)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Instruction Following14951378

Long Context Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 45.9 (#28), GPT-5.3 Chat: 42.6 (#120)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Longer Query14981396

Writing & Preference Claude Sonnet 5.5 leads

Claude Sonnet 5.5: 66.0 (#40), GPT-5.3 Chat: 63.1 (#68)

BenchmarkClaude Sonnet 5.5GPT-5.3 Chat
LMArena Text14711389
LMArena Creative Writing14651355
LMArena Multi-Turn14741412
EQ-Bench Creative Writing—1690

Frequently asked questions

Is Claude Sonnet 5.5 better than GPT-5.3 Chat?

Claude Sonnet 5.5 is the stronger model overall, scoring 61.9 to 42.8 on the Noometry Index.

Which is cheaper, Claude Sonnet 5.5 or GPT-5.3 Chat?

Claude Sonnet 5.5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-5.3 Chat lists at $1.75 and $14.

Is Claude Sonnet 5.5 or GPT-5.3 Chat better for coding?

Claude Sonnet 5.5 scores higher on coding benchmarks: 67.3 versus 41.4 in the Noometry coding category.

Which has the bigger context window?

Claude Sonnet 5.5 does, with 1M tokens against 128K.

How many benchmarks do Claude Sonnet 5.5 and GPT-5.3 Chat share?

12 benchmarks have published results for both models. Claude Sonnet 5.5 has 32 scored results on Noometry and GPT-5.3 Chat has 18.

Related comparisons

Go deeper