Noometry

Model comparison

DeepSeek-V3.1 vs Grok 4

Grok 4 is the stronger model overall, scoring 48.1 to 42.8 on the Noometry Index.

Last verified . 23 shared benchmarks.

DeepSeek-V3.1 DeepSeek

42.8

Rank #108 Confirmed

Grok 4 xAI

48.1

Rank #56 Confirmed

Summary

  • They share 23 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 1 category and Grok 4 in 7 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 leads 63.1 to 36.3.
  • The biggest single-benchmark swing is Fiction.LiveBench: 52.8% for DeepSeek-V3.1 and 94.4% for Grok 4.
  • DeepSeek-V3.1 has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.1Grok 4
ProviderDeepSeekxAI
Noometry Index42.848.1
Released2025-08-212025-07-09
WeightsOpenProprietary
Context window164K—
Max output8K—
Input $ / M tokens$0.25—
Output $ / M tokens$0.95—
Results tracked2748

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

Category by category

Coding Grok 4 leads

DeepSeek-V3.1: 40.3 (#144), Grok 4: 50.3 (#46)

BenchmarkDeepSeek-V3.1Grok 4
WeirdML38.4%45.7%
LMArena Coding14171408
Aider Polyglot—79.6%

Agentic & Tool Use Not comparable

DeepSeek-V3.1: —, Grok 4: 32.3 (#68)

BenchmarkDeepSeek-V3.1Grok 4
Terminal-Bench—27.2%
Berkeley Function Calling Leaderboard—63%
GDPval—21.1%
Cybench—43%
DeepResearch Bench—47.3%
BALROG—43.6%
LMArena Search—1142
METR Time Horizons—66.6%

Reasoning Grok 4 leads

DeepSeek-V3.1: 27.9 (#110), Grok 4: 36.7 (#65)

BenchmarkDeepSeek-V3.1Grok 4
SimpleBench40%60.5%
Kagi LLM Benchmark53.2%73.6%
LMArena Hard Prompts14171409
Epoch Capabilities Index139.92146.44
ForecastBench5860.9
ARC-AGI-2—16%
ARC-AGI-1—66.7%
Chess Puzzles—28%
DTBench82.7%—
LMCA24.3%—

Math Grok 4 leads

DeepSeek-V3.1: 38.9 (#122), Grok 4: 48.4 (#64)

BenchmarkDeepSeek-V3.1Grok 4
LMArena Math14201422
OTIS Mock AIME 2024-2025—84%
Omni-MATH—60.3%
FrontierMath (Feb 2025 set)—19.7%
FrontierMath Tier 4 (v1)—2.1%

Knowledge Grok 4 leads

DeepSeek-V3.1: 43.7 (#90), Grok 4: 53.8 (#55)

BenchmarkDeepSeek-V3.1Grok 4
LMArena Expert14051415
GPQA Diamond—87%
MMLU-Pro—85.1%
Confabulations—12.4%
Vectara Hallucination Rate5.5%—
GPQA (HELM)—72.7%

Multimodal Not comparable

DeepSeek-V3.1: —, Grok 4: 33.7 (#94)

BenchmarkDeepSeek-V3.1Grok 4
LMArena Vision—1210
GeoBench—45%

Multilingual Too close to call

DeepSeek-V3.1: 51.6 (#106), Grok 4: 51.8 (#103)

BenchmarkDeepSeek-V3.1Grok 4
LMArena Non-English14001403
LMArena Chinese14691427
LMArena French14471418
LMArena German14111429
LMArena Japanese13781394
LMArena Korean13371377
LMArena Russian14051410
LMArena Spanish14311420

Instruction Following Grok 4 leads

DeepSeek-V3.1: 73.9 (#110), Grok 4: 79.2 (#5)

BenchmarkDeepSeek-V3.1Grok 4
LMArena Instruction Following14001387
IFEval—94.9%

Long Context Grok 4 leads

DeepSeek-V3.1: 36.3 (#232), Grok 4: 63.1 (#4)

BenchmarkDeepSeek-V3.1Grok 4
Fiction.LiveBench52.8%94.4%
LMArena Longer Query14221409

Writing & Preference DeepSeek-V3.1 leads

DeepSeek-V3.1: 60.3 (#98), Grok 4: 58.5 (#116)

BenchmarkDeepSeek-V3.1Grok 4
LMArena Text14201411
LMArena Creative Writing14011397
LMArena Multi-Turn14081416
Short-Story Creative Writing—76.9%
EQ-Bench Creative Writing1436—
WildBench—79.7%

Frequently asked questions

Is DeepSeek-V3.1 better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 42.8 on the Noometry Index.

Is DeepSeek-V3.1 or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 40.3 in the Noometry coding category.

How many benchmarks do DeepSeek-V3.1 and Grok 4 share?

23 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Grok 4 has 48.

Related comparisons

Go deeper