Noometry

Model comparison

Kimi K3 vs Nova 2.0 Pro Preview

Kimi K3 is the stronger model overall, scoring 59.5 to 33.4 on the Noometry Index.

Last verified . 3 shared benchmarks.

Kimi K3 Moonshot AI

59.5

Rank #15 Confirmed

Nova 2.0 Pro Preview Amazon

33.4

Rank #244 Reported

Summary

  • They share 3 benchmarks with published results for both. Kimi K3 scores higher in 3 categories and Nova 2.0 Pro Preview in 0 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Kimi K3 leads 63.0 to 22.4.
  • The biggest single-benchmark swing is CritPt: 23.4% for Kimi K3 and 0% for Nova 2.0 Pro Preview.
  • Kimi K3 has downloadable open weights; the other is API-only.

Side by side

Kimi K3Nova 2.0 Pro Preview
ProviderMoonshot AIAmazon
Noometry Index59.533.4
Released2026-07-162025-12-02
WeightsOpenProprietary
Context window1.05M—
Max output1.05M—
Input $ / M tokens$3—
Output $ / M tokens$15—
Results tracked533

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

Category by category

Coding Kimi K3 leads

Kimi K3: 61.0 (#10), Nova 2.0 Pro Preview: 40.8 (#133)

BenchmarkKimi K3Nova 2.0 Pro Preview
SciCode59.5%42.7%
DeepSWE68.5%—
FrontierCode44.2%—
LMArena WebDev1654—
FrontierSWE25.9%—
WeirdML82.6%—
LMArena Coding1508—
ALE-Bench1,524—

Agentic & Tool Use Kimi K3 leads

Kimi K3: 41.8 (#20), Nova 2.0 Pro Preview: 21.9 (#136)

BenchmarkKimi K3Nova 2.0 Pro Preview
GDP.pdf19%2%
APEX-Agents50.6%—
τ²-bench Banking37.1%—
PostTrainBench32%—
GBAEval48.3%—
Vending-Bench 25,165—

Reasoning Kimi K3 leads

Kimi K3: 63.0 (#17), Nova 2.0 Pro Preview: 22.4 (#194)

BenchmarkKimi K3Nova 2.0 Pro Preview
CritPt23.4%0%
ARC-AGI-260.4%—
SimpleBench60.7%—
NYT Connections (extended)93.6%—
ARC-AGI-194.5%—
Chess Puzzles39%—
LMArena Hard Prompts1496—
Mystery Game Puzzles26%—
DTBench91.2%—
LMCA52.7%—
Surface Evolver Bench95%—
Epoch Capabilities Index157.45—
ForecastBench61.1—

Math Not comparable

Kimi K3: 74.2 (#16), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
FrontierMath (Tiers 1-3)72.2%—
FrontierMath Tier 439%—
MathArena Final-Answer Competitions87.8%—
OTIS Mock AIME 2024-202597.2%—
ProofBench87%—
LMArena Math1491—

Knowledge Not comparable

Kimi K3: 63.2 (#21), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
GPQA Diamond93.1%—
SimpleQA Verified50.6%—
LMArena Expert1521—

Multimodal Not comparable

Kimi K3: 37.8 (#70), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
Blueprint-Bench 229.5%—
Furniture Assembly34.2%—

Multilingual Not comparable

Kimi K3: 56.3 (#21), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
LMArena Non-English1466—
LMArena Chinese1529—
LMArena French1491—
LMArena German1488—
LMArena Japanese1487—
LMArena Korean1458—
LMArena Russian1482—
LMArena Spanish1472—

Instruction Following Not comparable

Kimi K3: 77.7 (#14), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
LMArena Instruction Following1483—

Long Context Not comparable

Kimi K3: 45.8 (#29), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
LMArena Longer Query1494—

Writing & Preference Not comparable

Kimi K3: 76.6 (#4), Nova 2.0 Pro Preview: —

BenchmarkKimi K3Nova 2.0 Pro Preview
LMArena Text1476—
LMArena Creative Writing1454—
EQ-Bench Creative Writing2082—
EQ-Bench 41339—
LMArena Multi-Turn1488—

Frequently asked questions

Is Kimi K3 better than Nova 2.0 Pro Preview?

Kimi K3 is the stronger model overall, scoring 59.5 to 33.4 on the Noometry Index.

Is Kimi K3 or Nova 2.0 Pro Preview better for coding?

Kimi K3 scores higher on coding benchmarks: 61.0 versus 40.8 in the Noometry coding category.

How many benchmarks do Kimi K3 and Nova 2.0 Pro Preview share?

3 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and Nova 2.0 Pro Preview has 3.

Related comparisons

Go deeper