Noometry

Model comparison

GPT-4 vs Kimi K2 Thinking Turbo

Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 29.1 on the Noometry Index.

Last verified . 20 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Kimi K2 Thinking Turbo Moonshot AI

45.8

Rank #70 Confirmed

Summary

  • They share 20 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Kimi K2 Thinking Turbo in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Kimi K2 Thinking Turbo leads 47.4 to 10.8.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 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-4Kimi K2 Thinking Turbo
ProviderOpenAIMoonshot AI
Noometry Index29.145.8
Released2023-03-142025-11-06
WeightsProprietaryOpen
Context window8K—
Max output8K—
Input $ / M tokens$30—
Output $ / M tokens$60—
Results tracked3821

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

Category by category

Coding Kimi K2 Thinking Turbo leads

GPT-4: 31.6 (#283), Kimi K2 Thinking Turbo: 38.4 (#178)

BenchmarkGPT-4Kimi K2 Thinking Turbo
LMArena Coding12541454
LMArena WebDev—1322
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Kimi K2 Thinking Turbo: —

BenchmarkGPT-4Kimi K2 Thinking Turbo
METR Time Horizons36.1%—

Reasoning Kimi K2 Thinking Turbo leads

GPT-4: 17.8 (#289), Kimi K2 Thinking Turbo: 32.5 (#75)

BenchmarkGPT-4Kimi K2 Thinking Turbo
Chess Puzzles4%20%
LMArena Hard Prompts12411428
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.89—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Kimi K2 Thinking Turbo leads

GPT-4: 10.8 (#309), Kimi K2 Thinking Turbo: 47.4 (#68)

BenchmarkGPT-4Kimi K2 Thinking Turbo
OTIS Mock AIME 2024-20251.1%83.1%
LMArena Math12691429
MATH Level 523%—
GSM8K92%—

Knowledge Kimi K2 Thinking Turbo leads

GPT-4: 18.4 (#282), Kimi K2 Thinking Turbo: 50.9 (#69)

BenchmarkGPT-4Kimi K2 Thinking Turbo
GPQA Diamond35.7%84.2%
LMArena Expert12111439
MMLU86.4%—
TriviaQA84.8%—

Multilingual Kimi K2 Thinking Turbo leads

GPT-4: 40.6 (#215), Kimi K2 Thinking Turbo: 51.4 (#109)

BenchmarkGPT-4Kimi K2 Thinking Turbo
LMArena Non-English12461398
LMArena Chinese12421456
LMArena French12831425
LMArena German12511390
LMArena Japanese12091357
LMArena Korean11841330
LMArena Russian12511391
LMArena Spanish12611406

Instruction Following Kimi K2 Thinking Turbo leads

GPT-4: 65.3 (#222), Kimi K2 Thinking Turbo: 74.0 (#109)

BenchmarkGPT-4Kimi K2 Thinking Turbo
LMArena Instruction Following12411403

Long Context Kimi K2 Thinking Turbo leads

GPT-4: 37.7 (#212), Kimi K2 Thinking Turbo: 43.2 (#102)

BenchmarkGPT-4Kimi K2 Thinking Turbo
LMArena Longer Query12441415

Writing & Preference Kimi K2 Thinking Turbo leads

GPT-4: 34.9 (#268), Kimi K2 Thinking Turbo: 60.0 (#104)

BenchmarkGPT-4Kimi K2 Thinking Turbo
LMArena Text12631415
LMArena Creative Writing12441374
LMArena Multi-Turn12571414
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than Kimi K2 Thinking Turbo?

Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 29.1 on the Noometry Index.

Is GPT-4 or Kimi K2 Thinking Turbo better for coding?

Kimi K2 Thinking Turbo scores higher on coding benchmarks: 38.4 versus 31.6 in the Noometry coding category.

How many benchmarks do GPT-4 and Kimi K2 Thinking Turbo share?

20 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Kimi K2 Thinking Turbo has 21.

Related comparisons

Go deeper