Noometry

Model comparison

DeepSeek V4 Flash vs Kimi K2.5 Instant

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.6 on the Noometry Index.

Last verified . 17 shared benchmarks.

DeepSeek V4 Flash DeepSeek

53.6

Rank #35 Confirmed

Kimi K2.5 Instant Moonshot AI

43.6

Rank #89 Confirmed

Summary

  • They share 17 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 6 categories and Kimi K2.5 Instant in 2 categories; 5 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 29.7.

Side by side

DeepSeek V4 FlashKimi K2.5 Instant
ProviderDeepSeekMoonshot AI
Noometry Index53.643.6
Released2026-04-24—
WeightsOpenOpen
Context window1M—
Max output393K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked4118

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

Category by category

Coding DeepSeek V4 Flash leads

DeepSeek V4 Flash: 47.9 (#59), Kimi K2.5 Instant: 42.6 (#97)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena WebDev15821404
LMArena Coding14571484
FrontierCode18.8%—
SciCode49.9%—
WeirdML63%—
ALE-Bench1,306—

Reasoning DeepSeek V4 Flash leads

DeepSeek V4 Flash: 53.7 (#30), Kimi K2.5 Instant: 29.7 (#90)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Hard Prompts14441443
ARC-AGI-261.4%—
SimpleBench61.1%—
Kagi LLM Benchmark52.2%—
NYT Connections (extended)89.6%—
ARC-AGI-189%—
CritPt16.6%—
Chess Puzzles33%—
Mystery Game Puzzles34%—
DTBench90.9%—
LMCA41.7%—
Epoch Capabilities Index154.49—

Math DeepSeek V4 Flash leads

DeepSeek V4 Flash: 60.3 (#37), Kimi K2.5 Instant: 39.4 (#105)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Math14271442
FrontierMath (Tiers 1-3)57.5%—
FrontierMath Tier 424.4%—
MathArena Final-Answer Competitions76.5%—
OTIS Mock AIME 2024-202594.4%—
ProofBench56%—

Knowledge DeepSeek V4 Flash leads

DeepSeek V4 Flash: 55.4 (#48), Kimi K2.5 Instant: 40.2 (#123)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Expert14411440
GPQA Diamond91%—
SimpleQA Verified33.6%—

Multimodal Not comparable

DeepSeek V4 Flash: —, Kimi K2.5 Instant: 40.2 (#50)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Vision—1254

Multilingual Too close to call

DeepSeek V4 Flash: 53.0 (#72), Kimi K2.5 Instant: 52.0 (#94)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Non-English14201406
LMArena Chinese14681449
LMArena French14391403
LMArena German14181413
LMArena Korean13841378
LMArena Russian14281404
LMArena Spanish14361447
LMArena Japanese1406—

Instruction Following Too close to call

DeepSeek V4 Flash: 74.9 (#81), Kimi K2.5 Instant: 75.3 (#65)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Instruction Following14211430

Long Context Too close to call

DeepSeek V4 Flash: 43.8 (#85), Kimi K2.5 Instant: 43.9 (#83)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Longer Query14341435

Writing & Preference DeepSeek V4 Flash leads

DeepSeek V4 Flash: 63.8 (#61), Kimi K2.5 Instant: 60.6 (#95)

BenchmarkDeepSeek V4 FlashKimi K2.5 Instant
LMArena Text14321420
LMArena Creative Writing14031381
LMArena Multi-Turn14491427
EQ-Bench Creative Writing1559—

Frequently asked questions

Is DeepSeek V4 Flash better than Kimi K2.5 Instant?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 43.6 on the Noometry Index.

Is DeepSeek V4 Flash or Kimi K2.5 Instant better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 42.6 in the Noometry coding category.

How many benchmarks do DeepSeek V4 Flash and Kimi K2.5 Instant share?

17 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Kimi K2.5 Instant has 18.

Related comparisons

Go deeper