Noometry

Model comparison

GPT-4.5 vs Kimi K3

Kimi K3 is the stronger model overall, scoring 59.5 to 37.2 on the Noometry Index.

Last verified . 25 shared benchmarks.

GPT-4.5 OpenAI

37.2

Rank #208 Confirmed

Kimi K3 Moonshot AI

59.5

Rank #15 Confirmed

Summary

  • They share 25 benchmarks with published results for both. GPT-4.5 scores higher in 0 categories and Kimi K3 in 10 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Kimi K3 leads 63.0 to 13.9.
  • The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 94.5% for Kimi K3.
  • Kimi K3 has downloadable open weights; the other is API-only.

Side by side

GPT-4.5Kimi K3
ProviderOpenAIMoonshot AI
Noometry Index37.259.5
Released2025-02-272026-07-16
WeightsProprietaryOpen
Context window—1.05M
Max output—1.05M
Input $ / M tokens—$3
Output $ / M tokens—$15
Results tracked4253

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

Category by category

Coding Kimi K3 leads

GPT-4.5: 42.2 (#109), Kimi K3: 61.0 (#10)

BenchmarkGPT-4.5Kimi K3
WeirdML39.4%82.6%
LMArena Coding13961508
DeepSWE—68.5%
FrontierCode—44.2%
Aider Polyglot44.9%—
LMArena WebDev—1654
FrontierSWE—25.9%
SciCode—59.5%
LiveBench Coding75.2%—
ALE-Bench—1,524

Agentic & Tool Use Kimi K3 leads

GPT-4.5: 27.9 (#97), Kimi K3: 41.8 (#20)

BenchmarkGPT-4.5Kimi K3
APEX-Agents—50.6%
τ²-bench Banking—37.1%
Cybench17.5%—
PostTrainBench—32%
GBAEval—48.3%
GDP.pdf—19%
Vending-Bench 2—5,165

Reasoning Kimi K3 leads

GPT-4.5: 13.9 (#330), Kimi K3: 63.0 (#17)

BenchmarkGPT-4.5Kimi K3
ARC-AGI-20.8%60.4%
SimpleBench34.5%60.7%
ARC-AGI-110.3%94.5%
LMArena Hard Prompts14031496
Epoch Capabilities Index136.74157.45
ForecastBench61.761.1
NYT Connections (extended)—93.6%
CritPt—23.4%
Chess Puzzles—39%
EnigmaEval3.2%—
LiveBench Reasoning71.1%—
Mystery Game Puzzles—26%
DTBench—91.2%
LiveBench Data Analysis64.3%—
LMCA—52.7%
Surface Evolver Bench—95%
LiveBench69%—

Math Kimi K3 leads

GPT-4.5: 32.6 (#211), Kimi K3: 74.2 (#16)

BenchmarkGPT-4.5Kimi K3
OTIS Mock AIME 2024-202537.8%97.2%
LMArena Math14121491
FrontierMath (Tiers 1-3)—72.2%
FrontierMath Tier 4—39%
MathArena Final-Answer Competitions—87.8%
ProofBench—87%
LiveBench Math69.3%—
MATH Level 578.6%—

Knowledge Kimi K3 leads

GPT-4.5: 32.5 (#211), Kimi K3: 63.2 (#21)

BenchmarkGPT-4.5Kimi K3
GPQA Diamond68.7%93.1%
LMArena Expert13941521
Humanity's Last Exam5.4%—
SimpleQA Verified—50.6%
Confabulations13.6%—

Multimodal Too close to call

GPT-4.5: 37.6 (#71), Kimi K3: 37.8 (#70)

BenchmarkGPT-4.5Kimi K3
LMArena Vision1195—
VPCT45%—
Blueprint-Bench 2—29.5%
Furniture Assembly—34.2%

Multilingual Kimi K3 leads

GPT-4.5: 52.5 (#83), Kimi K3: 56.3 (#21)

BenchmarkGPT-4.5Kimi K3
LMArena Non-English14131466
LMArena Chinese14211529
LMArena French14181491
LMArena German14571488
LMArena Japanese14161487
LMArena Korean13921458
LMArena Russian14191482
LMArena Spanish—1472

Instruction Following Kimi K3 leads

GPT-4.5: 72.6 (#134), Kimi K3: 77.7 (#14)

BenchmarkGPT-4.5Kimi K3
LMArena Instruction Following14041483
LiveBench Instruction Following72.3%—

Long Context Kimi K3 leads

GPT-4.5: 40.4 (#155), Kimi K3: 45.8 (#29)

BenchmarkGPT-4.5Kimi K3
LMArena Longer Query14061494
Fiction.LiveBench63.9%—

Writing & Preference Kimi K3 leads

GPT-4.5: 56.9 (#134), Kimi K3: 76.6 (#4)

BenchmarkGPT-4.5Kimi K3
LMArena Text14171476
LMArena Creative Writing13941454
EQ-Bench Creative Writing12582082
LMArena Multi-Turn14441488
Short-Story Creative Writing75.6%—
EQ-Bench 4—1339
LiveBench Language61.5%—

Frequently asked questions

Is GPT-4.5 better than Kimi K3?

Kimi K3 is the stronger model overall, scoring 59.5 to 37.2 on the Noometry Index.

Is GPT-4.5 or Kimi K3 better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 42.2 in the Noometry coding category.

How many benchmarks do GPT-4.5 and Kimi K3 share?

25 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Kimi K3 has 53.

Related comparisons

Go deeper