Noometry

Model comparison

Deepseek Coder v2 vs Grok 4

Grok 4 is the stronger model overall, scoring 48.1 to 35.9 on the Noometry Index.

Last verified . 17 shared benchmarks.

Deepseek Coder v2 DeepSeek

35.9

Rank #220 Confirmed

Grok 4 xAI

48.1

Rank #56 Confirmed

Summary

  • They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 0 categories and Grok 4 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in long context, where Grok 4 leads 63.1 to 37.0.
  • Deepseek Coder v2 has downloadable open weights; the other is API-only.

Side by side

Deepseek Coder v2Grok 4
ProviderDeepSeekxAI
Noometry Index35.948.1
Released2024-06-172025-07-09
WeightsOpenProprietary
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked2448

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

Category by category

Coding Grok 4 leads

Deepseek Coder v2: 38.1 (#183), Grok 4: 50.3 (#46)

BenchmarkDeepseek Coder v2Grok 4
LMArena Coding12511408
Aider Polyglot—79.6%
WeirdML—45.7%
BigCodeBench Instruct48.2%—
BigCodeBench Complete59.7%—
HumanEval+82.3%—
MBPP+75.1%—

Agentic & Tool Use Not comparable

Deepseek Coder v2: —, Grok 4: 32.3 (#68)

BenchmarkDeepseek Coder v2Grok 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 Coder v2: 23.6 (#176), Grok 4: 36.7 (#65)

BenchmarkDeepseek Coder v2Grok 4
LMArena Hard Prompts12071409
ARC-AGI-2—16%
SimpleBench—60.5%
Kagi LLM Benchmark—73.6%
ARC-AGI-1—66.7%
Chess Puzzles—28%
Epoch Capabilities Index—146.44
ForecastBench—60.9
WinoGrande83.7%—

Math Grok 4 leads

Deepseek Coder v2: 34.9 (#190), Grok 4: 48.4 (#64)

BenchmarkDeepseek Coder v2Grok 4
LMArena Math12411422
OTIS Mock AIME 2024-2025—84%
Omni-MATH—60.3%
FrontierMath (Feb 2025 set)—19.7%
FrontierMath Tier 4 (v1)—2.1%
GSM8K94.5%—

Knowledge Grok 4 leads

Deepseek Coder v2: 32.3 (#212), Grok 4: 53.8 (#55)

BenchmarkDeepseek Coder v2Grok 4
LMArena Expert11811415
GPQA Diamond—87%
MMLU-Pro—85.1%
Confabulations—12.4%
GPQA (HELM)—72.7%
ARC (AI2) Challenge64.3%—

Multimodal Not comparable

Deepseek Coder v2: —, Grok 4: 33.7 (#94)

BenchmarkDeepseek Coder v2Grok 4
LMArena Vision—1210
GeoBench—45%

Multilingual Grok 4 leads

Deepseek Coder v2: 36.3 (#240), Grok 4: 51.8 (#103)

BenchmarkDeepseek Coder v2Grok 4
LMArena Non-English11821403
LMArena Chinese12011427
LMArena French11851418
LMArena German11641429
LMArena Japanese11261394
LMArena Korean11041377
LMArena Russian11881410
LMArena Spanish11531420

Instruction Following Grok 4 leads

Deepseek Coder v2: 61.7 (#242), Grok 4: 79.2 (#5)

BenchmarkDeepseek Coder v2Grok 4
LMArena Instruction Following11801387
IFEval—94.9%

Long Context Grok 4 leads

Deepseek Coder v2: 37.0 (#224), Grok 4: 63.1 (#4)

BenchmarkDeepseek Coder v2Grok 4
LMArena Longer Query12191409
Fiction.LiveBench—94.4%

Writing & Preference Grok 4 leads

Deepseek Coder v2: 38.2 (#253), Grok 4: 58.5 (#116)

BenchmarkDeepseek Coder v2Grok 4
LMArena Text11911411
LMArena Creative Writing11201397
LMArena Multi-Turn11771416
Short-Story Creative Writing—76.9%
WildBench—79.7%

Frequently asked questions

Is Deepseek Coder v2 better than Grok 4?

Grok 4 is the stronger model overall, scoring 48.1 to 35.9 on the Noometry Index.

Is Deepseek Coder v2 or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 50.3 versus 38.1 in the Noometry coding category.

How many benchmarks do Deepseek Coder v2 and Grok 4 share?

17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Grok 4 has 48.

Related comparisons

Go deeper