Noometry

Model comparison

Claude Sonnet 5 vs GPT-5.2 Pro

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 52.3 on the Noometry Index.

Last verified . 5 shared benchmarks.

Claude Sonnet 5 Anthropic

54.6

Rank #29 Confirmed

GPT-5.2 Pro OpenAI

52.3

Rank #39 Reported

Summary

  • They share 5 benchmarks with published results for both. Claude Sonnet 5 scores higher in 1 category and GPT-5.2 Pro in 1 category; one gap is clear of the uncertainty.
  • The biggest single-benchmark swing is FrontierMath Tier 4: 29.3% for Claude Sonnet 5 and 46% for GPT-5.2 Pro.
  • Claude Sonnet 5 is cheaper at $2 / $10 per million input/output tokens, against $21 / $168 for GPT-5.2 Pro.
  • Claude Sonnet 5 accepts more context: 1M tokens versus 400K.

Side by side

Claude Sonnet 5GPT-5.2 Pro
ProviderAnthropicOpenAI
Noometry Index54.652.3
Released2026-06-292025-12-11
WeightsProprietaryProprietary
Context window1M400K
Max output128K128K
Input $ / M tokens$2$21
Output $ / M tokens$10$168
Results tracked518

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

Category by category

Coding Not comparable

Claude Sonnet 5: 55.5 (#26), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
DeepSWE53.8%—
FrontierCode42.7%—
CursorBench34.1%—
LMArena WebDev1541—
SciCode54.3%—
GSO37.3%—
WeirdML68.8%—
LMArena Coding1483—
ALE-Bench1,463—

Agentic & Tool Use Not comparable

Claude Sonnet 5: 42.8 (#18), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
APEX-Agents54.5%—
GBAEval65.3%—
LMArena Search1194—
Vending-Bench 26,378—

Reasoning GPT-5.2 Pro leads

Claude Sonnet 5: 49.1 (#39), GPT-5.2 Pro: 51.5 (#33)

BenchmarkClaude Sonnet 5GPT-5.2 Pro
SimpleBench60.6%57.4%
NYT Connections (extended)75.1%79.3%
Epoch Capabilities Index156.21155.4
ARC-AGI-2—54.2%
ARC-AGI-1—90.5%
CritPt16.9%—
Chess Puzzles35%—
LMArena Hard Prompts1461—
Mystery Game Puzzles35%—
DTBench92.5%—
LMCA50%—
Surface Evolver Bench60%—
Bench to the Future 30.14—
ForecastBench61.1—

Math Too close to call

Claude Sonnet 5: 66.2 (#27), GPT-5.2 Pro: 65.3 (#29)

BenchmarkClaude Sonnet 5GPT-5.2 Pro
FrontierMath (Tiers 1-3)65.6%74%
FrontierMath Tier 429.3%46%
OTIS Mock AIME 2024-202594.7%—
ProofBench77%—
LMArena Math1467—
FrontierMath Tier 4 (v1)—31.3%

Knowledge Not comparable

Claude Sonnet 5: 55.6 (#47), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
GPQA Diamond90.5%—
SimpleQA Verified33.7%—
LMArena Expert1490—

Multimodal Not comparable

Claude Sonnet 5: 42.4 (#31), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
LMArena Vision1274—
Blueprint-Bench 224.9%—
LMArena Document1466—

Multilingual Not comparable

Claude Sonnet 5: 53.8 (#55), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
LMArena Non-English1431—
LMArena Chinese1477—
LMArena French1460—
LMArena German1440—
LMArena Japanese1422—
LMArena Korean1411—
LMArena Russian1451—
LMArena Spanish1437—

Instruction Following Not comparable

Claude Sonnet 5: 76.3 (#41), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
LMArena Instruction Following1452—

Long Context Not comparable

Claude Sonnet 5: 44.8 (#55), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
LMArena Longer Query1463—

Writing & Preference Not comparable

Claude Sonnet 5: 69.2 (#25), GPT-5.2 Pro: —

BenchmarkClaude Sonnet 5GPT-5.2 Pro
LMArena Text1442—
LMArena Creative Writing1416—
EQ-Bench Creative Writing1794—
EQ-Bench 41236—
LMArena Multi-Turn1454—

Frequently asked questions

Is Claude Sonnet 5 better than GPT-5.2 Pro?

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 52.3 on the Noometry Index.

Which is cheaper, Claude Sonnet 5 or GPT-5.2 Pro?

Claude Sonnet 5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; GPT-5.2 Pro lists at $21 and $168.

Which has the bigger context window?

Claude Sonnet 5 does, with 1M tokens against 400K.

How many benchmarks do Claude Sonnet 5 and GPT-5.2 Pro share?

5 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and GPT-5.2 Pro has 8.

Related comparisons

Go deeper