Noometry

Model comparison

DeepSeek V4 Flash vs Grok-3 mini

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 41.2 on the Noometry Index.

Last verified . 24 shared benchmarks.

DeepSeek V4 Flash DeepSeek

53.6

Rank #35 Confirmed

Grok-3 mini xAI

41.2

Rank #141 Confirmed

Summary

  • They share 24 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 7 categories and Grok-3 mini in 1 category; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 13.6.
  • The biggest single-benchmark swing is ARC-AGI-1: 89% for DeepSeek V4 Flash and 16.5% for Grok-3 mini.
  • DeepSeek V4 Flash has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4 FlashGrok-3 mini
ProviderDeepSeekxAI
Noometry Index53.641.2
Released2026-04-242025-04-09
WeightsOpenProprietary
Context window1M—
Max output393K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked4135

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), Grok-3 mini: 40.8 (#131)

BenchmarkDeepSeek V4 FlashGrok-3 mini
WeirdML63%42.6%
LMArena Coding14571379
FrontierCode18.8%—
Aider Polyglot—49.3%
LMArena WebDev1582—
SciCode49.9%—
ALE-Bench1,306—

Reasoning DeepSeek V4 Flash leads

DeepSeek V4 Flash: 53.7 (#30), Grok-3 mini: 13.6 (#334)

BenchmarkDeepSeek V4 FlashGrok-3 mini
ARC-AGI-261.4%0.4%
Kagi LLM Benchmark52.2%61.3%
ARC-AGI-189%16.5%
LMArena Hard Prompts14441375
Epoch Capabilities Index154.49140.35
SimpleBench61.1%—
NYT Connections (extended)89.6%—
CritPt16.6%—
Chess Puzzles33%—
Mystery Game Puzzles34%—
DTBench90.9%—
LMCA41.7%—

Math DeepSeek V4 Flash leads

DeepSeek V4 Flash: 60.3 (#37), Grok-3 mini: 42.1 (#85)

BenchmarkDeepSeek V4 FlashGrok-3 mini
OTIS Mock AIME 2024-202594.4%77.8%
LMArena Math14271386
FrontierMath (Tiers 1-3)57.5%—
FrontierMath Tier 424.4%—
MathArena Final-Answer Competitions76.5%—
ProofBench56%—
Omni-MATH—31.8%
MATH Level 5—90.9%
FrontierMath (Feb 2025 set)—5.9%

Knowledge DeepSeek V4 Flash leads

DeepSeek V4 Flash: 55.4 (#48), Grok-3 mini: 46.4 (#81)

BenchmarkDeepSeek V4 FlashGrok-3 mini
GPQA Diamond91%76.3%
LMArena Expert14411395
SimpleQA Verified33.6%—
MMLU-Pro—79.9%
Confabulations—10.8%
GPQA (HELM)—67.5%

Multilingual DeepSeek V4 Flash leads

DeepSeek V4 Flash: 53.0 (#72), Grok-3 mini: 48.1 (#145)

BenchmarkDeepSeek V4 FlashGrok-3 mini
LMArena Non-English14201352
LMArena Chinese14681387
LMArena French14391357
LMArena German14181349
LMArena Japanese14061342
LMArena Korean13841335
LMArena Russian14281353
LMArena Spanish14361381

Instruction Following Grok-3 mini leads

DeepSeek V4 Flash: 74.9 (#81), Grok-3 mini: 78.5 (#9)

BenchmarkDeepSeek V4 FlashGrok-3 mini
LMArena Instruction Following14211357
IFEval—95.1%

Long Context DeepSeek V4 Flash leads

DeepSeek V4 Flash: 43.8 (#85), Grok-3 mini: 41.0 (#147)

BenchmarkDeepSeek V4 FlashGrok-3 mini
LMArena Longer Query14341372
Fiction.LiveBench—66.7%

Writing & Preference DeepSeek V4 Flash leads

DeepSeek V4 Flash: 63.8 (#61), Grok-3 mini: 52.5 (#169)

BenchmarkDeepSeek V4 FlashGrok-3 mini
LMArena Text14321370
LMArena Creative Writing14031342
LMArena Multi-Turn14491355
Short-Story Creative Writing—73.5%
EQ-Bench Creative Writing1559—
WildBench—65.1%

Frequently asked questions

Is DeepSeek V4 Flash better than Grok-3 mini?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 41.2 on the Noometry Index.

Is DeepSeek V4 Flash or Grok-3 mini better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 40.8 in the Noometry coding category.

How many benchmarks do DeepSeek V4 Flash and Grok-3 mini share?

24 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Grok-3 mini has 35.

Related comparisons

Go deeper