Noometry

Model comparison

Deepseek Coder v2 vs Grok-2 (Dec 2024)

Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 33.7 on the Noometry Index.

Last verified . 17 shared benchmarks.

Deepseek Coder v2 DeepSeek

35.9

Rank #220 Confirmed

Grok-2 (Dec 2024) xAI

33.7

Rank #239 Confirmed

Summary

  • They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 4 categories and Grok-2 (Dec 2024) in 4 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Deepseek Coder v2 leads 34.9 to 20.8.
  • Deepseek Coder v2 has downloadable open weights; the other is API-only.

Side by side

Deepseek Coder v2Grok-2 (Dec 2024)
ProviderDeepSeekxAI
Noometry Index35.933.7
Released2024-06-172024-08-13
WeightsOpenProprietary
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked2434

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

Category by category

Coding Deepseek Coder v2 leads

Deepseek Coder v2: 38.1 (#183), Grok-2 (Dec 2024): 33.3 (#258)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Coding12511287
WeirdML—22.2%
BigCodeBench Instruct48.2%—
LiveBench Coding—46.4%
BigCodeBench Complete59.7%—
HumanEval+82.3%—
MBPP+75.1%—

Reasoning Deepseek Coder v2 leads

Deepseek Coder v2: 23.6 (#176), Grok-2 (Dec 2024): 16.9 (#299)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Hard Prompts12071272
SimpleBench—22.7%
LiveBench Reasoning—54.8%
DTBench—65.2%
LiveBench Data Analysis—54.5%
Epoch Capabilities Index—130.48
LiveBench—54.3%
WinoGrande83.7%—

Math Deepseek Coder v2 leads

Deepseek Coder v2: 34.9 (#190), Grok-2 (Dec 2024): 20.8 (#284)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Math12411283
OTIS Mock AIME 2024-2025—11.5%
LiveBench Math—54.9%
MATH Level 5—63.5%
FrontierMath (Feb 2025 set)—0.7%
GSM8K94.5%—

Knowledge Deepseek Coder v2 leads

Deepseek Coder v2: 32.3 (#212), Grok-2 (Dec 2024): 29.8 (#233)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Expert11811254
GPQA Diamond—53.8%
Confabulations—20.1%
ARC (AI2) Challenge64.3%—

Multilingual Grok-2 (Dec 2024) leads

Deepseek Coder v2: 36.3 (#240), Grok-2 (Dec 2024): 43.1 (#188)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Non-English11821282
LMArena Chinese12011289
LMArena French11851318
LMArena German11641287
LMArena Japanese11261244
LMArena Korean11041237
LMArena Russian11881286
LMArena Spanish11531281

Instruction Following Grok-2 (Dec 2024) leads

Deepseek Coder v2: 61.7 (#242), Grok-2 (Dec 2024): 66.9 (#202)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Instruction Following11801270
LiveBench Instruction Following—69.6%

Long Context Grok-2 (Dec 2024) leads

Deepseek Coder v2: 37.0 (#224), Grok-2 (Dec 2024): 38.8 (#190)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Longer Query12191276

Writing & Preference Grok-2 (Dec 2024) leads

Deepseek Coder v2: 38.2 (#253), Grok-2 (Dec 2024): 48.6 (#198)

BenchmarkDeepseek Coder v2Grok-2 (Dec 2024)
LMArena Text11911305
LMArena Creative Writing11201284
LMArena Multi-Turn11771290
Short-Story Creative Writing—63.6%
LiveBench Language—45.6%

Frequently asked questions

Is Deepseek Coder v2 better than Grok-2 (Dec 2024)?

Deepseek Coder v2 is the stronger model overall, scoring 35.9 to 33.7 on the Noometry Index.

Is Deepseek Coder v2 or Grok-2 (Dec 2024) better for coding?

Deepseek Coder v2 scores higher on coding benchmarks: 38.1 versus 33.3 in the Noometry coding category.

How many benchmarks do Deepseek Coder v2 and Grok-2 (Dec 2024) share?

17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Grok-2 (Dec 2024) has 34.

Related comparisons

Go deeper