Noometry

Model comparison

GPT-4 vs Kimi K2.7 Code

Kimi K2.7 Code is the stronger model overall, scoring 43.0 to 28.9 on the Noometry Index.

Last verified . 5 shared benchmarks.

GPT-4 OpenAI

28.9

Rank #308 Confirmed

Kimi K2.7 Code Moonshot AI

43.0

Rank #90 Confirmed

Summary

  • They share 5 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Kimi K2.7 Code in 4 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in math, where Kimi K2.7 Code leads 52.6 to 10.1.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 95.6% for Kimi K2.7 Code.
  • Kimi K2.7 Code is cheaper at $0.95 / $4 per million input/output tokens, against $30 / $60 for GPT-4.
  • Kimi K2.7 Code accepts more context: 262K tokens versus 8K.
  • Kimi K2.7 Code has downloadable open weights; the other is API-only.

Side by side

GPT-4Kimi K2.7 Code
ProviderOpenAIMoonshot AI
Noometry Index28.943.0
Released2023-03-142026-06-12
WeightsProprietaryOpen
Context window8K262K
Max output8K262K
Input $ / M tokens$30$0.95
Output $ / M tokens$60$4
Results tracked3819

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

Category by category

Coding Kimi K2.7 Code leads

GPT-4: 31.2 (#277), Kimi K2.7 Code: 42.4 (#92)

BenchmarkGPT-4Kimi K2.7 Code
WeirdML12.4%54.1%
DeepSWE—30.5%
FrontierCode—30.1%
LMArena WebDev—1473
SciCode—47.5%
BigCodeBench Instruct46%—
LMArena Coding1254—
BigCodeBench Complete57.2%—
ALE-Bench—886.23
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Kimi K2.7 Code: 23.9 (#115)

BenchmarkGPT-4Kimi K2.7 Code
APEX-Agents—37.6%
GBAEval—0.9%
METR Time Horizons36.1%—
Vending-Bench 2—5,083

Reasoning Kimi K2.7 Code leads

GPT-4: 17.3 (#285), Kimi K2.7 Code: 38.2 (#56)

BenchmarkGPT-4Kimi K2.7 Code
Chess Puzzles4%21%
Epoch Capabilities Index125.88150.01
SimpleBench—57.9%
CritPt—10%
LMArena Hard Prompts1241—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
Surface Evolver Bench—48.8%
BIG-Bench Hard75.1%—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Kimi K2.7 Code leads

GPT-4: 10.1 (#304), Kimi K2.7 Code: 52.6 (#39)

BenchmarkGPT-4Kimi K2.7 Code
OTIS Mock AIME 2024-20251.1%95.6%
FrontierMath (Tiers 1-3)—54%
FrontierMath Tier 4—12.2%
LMArena Math1269—
MATH Level 523%—
GSM8K92%—

Knowledge Kimi K2.7 Code leads

GPT-4: 18.4 (#277), Kimi K2.7 Code: 53.6 (#51)

BenchmarkGPT-4Kimi K2.7 Code
GPQA Diamond35.7%87.9%
SimpleQA Verified—36.5%
LMArena Expert1211—
MMLU86.4%—
TriviaQA84.8%—

Multilingual Not comparable

GPT-4: 40.8 (#210), Kimi K2.7 Code: —

BenchmarkGPT-4Kimi K2.7 Code
LMArena Non-English1246—
LMArena Chinese1242—
LMArena French1282—
LMArena German1251—
LMArena Japanese1209—
LMArena Korean1184—
LMArena Russian1251—
LMArena Spanish1261—

Instruction Following Not comparable

GPT-4: 65.5 (#217), Kimi K2.7 Code: —

BenchmarkGPT-4Kimi K2.7 Code
LMArena Instruction Following1241—

Long Context Not comparable

GPT-4: 37.7 (#207), Kimi K2.7 Code: —

BenchmarkGPT-4Kimi K2.7 Code
LMArena Longer Query1244—

Writing & Preference Not comparable

GPT-4: 35.0 (#263), Kimi K2.7 Code: —

BenchmarkGPT-4Kimi K2.7 Code
LMArena Text1263—
LMArena Creative Writing1244—
EQ-Bench Creative Writing752—
LMArena Multi-Turn1257—

Frequently asked questions

Is GPT-4 better than Kimi K2.7 Code?

Kimi K2.7 Code is the stronger model overall, scoring 43.0 to 28.9 on the Noometry Index.

Which is cheaper, GPT-4 or Kimi K2.7 Code?

Kimi K2.7 Code is cheaper. It lists at $0.95 per million input tokens and $4 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or Kimi K2.7 Code better for coding?

Kimi K2.7 Code scores higher on coding benchmarks: 42.4 versus 31.2 in the Noometry coding category.

Which has the bigger context window?

Kimi K2.7 Code does, with 262K tokens against 8K.

How many benchmarks do GPT-4 and Kimi K2.7 Code share?

5 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Kimi K2.7 Code has 19.

Related comparisons

Go deeper