Noometry

Model comparison

Gemini 3.1 Pro Preview vs Kimi K2.7 Code

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 43.0 on the Noometry Index. Kimi K2.7 Code costs 2.6× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Last verified . 17 shared benchmarks.

Gemini 3.1 Pro Preview Google

56.7

Rank #20 Confirmed

Kimi K2.7 Code Moonshot AI

43.0

Rank #90 Confirmed

Summary

  • They share 17 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 5 categories and Kimi K2.7 Code in 0 categories; 4 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 38.2.
  • The biggest single-benchmark swing is SimpleQA Verified: 73.5% for Gemini 3.1 Pro Preview and 36.5% for Kimi K2.7 Code.
  • Kimi K2.7 Code is cheaper at $0.95 / $4 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 262K.
  • Kimi K2.7 Code has downloadable open weights; the other is API-only.

Side by side

Gemini 3.1 Pro PreviewKimi K2.7 Code
ProviderGoogleMoonshot AI
Noometry Index56.743.0
Released2026-02-192026-06-12
WeightsProprietaryOpen
Context window1.05M262K
Max output66K262K
Input $ / M tokens$2$0.95
Output $ / M tokens$12$4
Results tracked7119

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

Category by category

Coding Too close to call

Gemini 3.1 Pro Preview: 42.7 (#88), Kimi K2.7 Code: 42.4 (#92)

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
DeepSWE11.7%30.5%
LMArena WebDev14461473
SciCode58.9%47.5%
WeirdML72.1%54.1%
ALE-Bench1,161886.23
SWE-bench Verified75.6%—
FrontierCode—30.1%
GSO22.6%—
LMArena Coding1483—
MirrorCode8.9%—
AlgoTune2.02—

Agentic & Tool Use Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 37.7 (#33), Kimi K2.7 Code: 23.9 (#115)

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
APEX-Agents35.3%37.6%
GBAEval0.8%0.9%
Vending-Bench 23,7745,083
Terminal-Bench80.2%—
τ²-bench Banking26%—
DeepResearch Bench47.8%—
PostTrainBench22%—
BALROG57%—
ExploitBench26.1%—
GDP.pdf17%—
LMArena Search1211—
METR Time Horizons77%—

Reasoning Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.7 (#9), Kimi K2.7 Code: 38.2 (#56)

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
SimpleBench79.6%57.9%
CritPt17.7%10%
Chess Puzzles55%21%
Epoch Capabilities Index154.92150.01
ARC-AGI-277.1%—
NYT Connections (extended)97.4%—
ARC-AGI-198%—
EnigmaEval36.8%—
Thematic Generalization79.4%—
EBR-Bench14.3%—
LMArena Hard Prompts1484—
Mystery Game Puzzles34%—
DTBench97.1%—
LMCA53.8%—
Surface Evolver Bench—48.8%
ForecastBench59—

Math Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 61.8 (#28), Kimi K2.7 Code: 52.6 (#39)

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
FrontierMath (Tiers 1-3)59.6%54%
FrontierMath Tier 426.8%12.2%
OTIS Mock AIME 2024-202595.6%95.6%
MathArena Final-Answer Competitions86.5%—
ProofBench26%—
LMArena Math1485—
FrontierMath (Feb 2025 set)36.9%—
FrontierMath Tier 4 (v1)16.7%—

Knowledge Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.8 (#3), Kimi K2.7 Code: 53.6 (#51)

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
GPQA Diamond94.4%87.9%
SimpleQA Verified73.5%36.5%
Humanity's Last Exam46.4%—
Vectara Hallucination Rate10.4%—
LMArena Expert1484—

Multimodal Not comparable

Gemini 3.1 Pro Preview: 38.1 (#60), Kimi K2.7 Code: —

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
LMArena Vision1296—
Blueprint-Bench 226.5%—
Furniture Assembly26.7%—
LMArena Document1444—

Multilingual Not comparable

Gemini 3.1 Pro Preview: 57.1 (#11), Kimi K2.7 Code: —

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
LMArena Non-English1475—
LMArena Chinese1528—
LMArena French1488—
LMArena German1491—
LMArena Japanese1494—
LMArena Korean1458—
LMArena Russian1496—
LMArena Spanish1480—

Instruction Following Not comparable

Gemini 3.1 Pro Preview: 77.1 (#31), Kimi K2.7 Code: —

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
LMArena Instruction Following1465—

Long Context Not comparable

Gemini 3.1 Pro Preview: 47.3 (#17), Kimi K2.7 Code: —

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
CL-bench20.8%—
CL-bench Life16.9%—
LMArena Longer Query1482—

Writing & Preference Not comparable

Gemini 3.1 Pro Preview: 66.3 (#36), Kimi K2.7 Code: —

BenchmarkGemini 3.1 Pro PreviewKimi K2.7 Code
LMArena Text1480—
LMArena Creative Writing1482—
EQ-Bench Creative Writing1491—
EQ-Bench 41142—
LMArena Multi-Turn1486—

Frequently asked questions

Is Gemini 3.1 Pro Preview better than Kimi K2.7 Code?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 43.0 on the Noometry Index. Kimi K2.7 Code costs 2.6× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Which is cheaper, Gemini 3.1 Pro Preview 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; Gemini 3.1 Pro Preview lists at $2 and $12.

Is Gemini 3.1 Pro Preview or Kimi K2.7 Code better for coding?

They score almost the same on coding (42.7 vs 42.4); test both on your own repository before choosing.

Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 262K.

How many benchmarks do Gemini 3.1 Pro Preview and Kimi K2.7 Code share?

17 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Kimi K2.7 Code has 19.

Related comparisons

Go deeper