Noometry

Model comparison

GPT-4.5 vs Kimi K2 Thinking Turbo

Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.4 to 36.9 on the Noometry Index.

Last verified . 18 shared benchmarks.

GPT-4.5 OpenAI

36.9

Rank #200 Confirmed

Kimi K2 Thinking Turbo Moonshot AI

45.4

Rank #63 Confirmed

Summary

  • They share 18 benchmarks with published results for both. GPT-4.5 scores higher in 2 categories and Kimi K2 Thinking Turbo in 6 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Kimi K2 Thinking Turbo leads 50.9 to 32.4.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 83.1% for Kimi K2 Thinking Turbo.
  • Kimi K2 Thinking Turbo has downloadable open weights; the other is API-only.

Side by side

GPT-4.5Kimi K2 Thinking Turbo
ProviderOpenAIMoonshot AI
Noometry Index36.945.4
Released2025-02-272025-11-06
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked4221

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

Category by category

Coding GPT-4.5 leads

GPT-4.5: 41.8 (#102), Kimi K2 Thinking Turbo: 38.1 (#166)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Coding13971454
Aider Polyglot44.9%—
LMArena WebDev—1322
WeirdML39.4%—
LiveBench Coding75.2%—

Agentic & Tool Use Not comparable

GPT-4.5: 27.6 (#92), Kimi K2 Thinking Turbo: —

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
Cybench17.5%—

Reasoning Kimi K2 Thinking Turbo leads

GPT-4.5: 14.1 (#320), Kimi K2 Thinking Turbo: 31.4 (#72)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Hard Prompts14041428
ARC-AGI-20.8%—
SimpleBench34.5%—
ARC-AGI-110.3%—
Chess Puzzles—20%
EnigmaEval3.2%—
LiveBench Reasoning71.1%—
LiveBench Data Analysis64.3%—
Epoch Capabilities Index136.74—
ForecastBench61.7—
LiveBench69%—

Math Kimi K2 Thinking Turbo leads

GPT-4.5: 31.0 (#200), Kimi K2 Thinking Turbo: 45.5 (#62)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
OTIS Mock AIME 2024-202537.8%83.1%
LMArena Math14121429
LiveBench Math69.3%—
MATH Level 578.6%—

Knowledge Kimi K2 Thinking Turbo leads

GPT-4.5: 32.4 (#205), Kimi K2 Thinking Turbo: 50.9 (#64)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
GPQA Diamond68.7%84.2%
LMArena Expert13941439
Humanity's Last Exam5.4%—
Confabulations13.6%—

Multimodal Not comparable

GPT-4.5: 37.0 (#68), Kimi K2 Thinking Turbo: —

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Vision1195—
VPCT45%—

Multilingual GPT-4.5 leads

GPT-4.5: 52.7 (#78), Kimi K2 Thinking Turbo: 51.6 (#104)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Non-English14131398
LMArena Chinese14211457
LMArena French14171425
LMArena German14571390
LMArena Japanese14161356
LMArena Korean13921329
LMArena Russian14191390
LMArena Spanish—1407

Instruction Following Kimi K2 Thinking Turbo leads

GPT-4.5: 72.8 (#129), Kimi K2 Thinking Turbo: 74.2 (#103)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Instruction Following14041403
LiveBench Instruction Following72.3%—

Long Context Kimi K2 Thinking Turbo leads

GPT-4.5: 40.4 (#150), Kimi K2 Thinking Turbo: 43.2 (#99)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Longer Query14061415
Fiction.LiveBench63.9%—

Writing & Preference Kimi K2 Thinking Turbo leads

GPT-4.5: 57.1 (#129), Kimi K2 Thinking Turbo: 60.2 (#99)

BenchmarkGPT-4.5Kimi K2 Thinking Turbo
LMArena Text14171415
LMArena Creative Writing13941373
LMArena Multi-Turn14441414
Short-Story Creative Writing75.6%—
EQ-Bench Creative Writing1258—
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than Kimi K2 Thinking Turbo?

Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.4 to 36.9 on the Noometry Index.

Is GPT-4.5 or Kimi K2 Thinking Turbo better for coding?

GPT-4.5 scores higher on coding benchmarks: 41.8 versus 38.1 in the Noometry coding category.

How many benchmarks do GPT-4.5 and Kimi K2 Thinking Turbo share?

18 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Kimi K2 Thinking Turbo has 21.

Related comparisons

Go deeper