Noometry

Model comparison

Deepseek Coder v2 vs Gemma 4 31B IT

Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 35.9 on the Noometry Index.

Last verified . 14 shared benchmarks.

Deepseek Coder v2 DeepSeek

35.9

Rank #220 Confirmed

Gemma 4 31B IT Google

43.5

Rank #90 Confirmed

Summary

  • They share 14 benchmarks with published results for both. Deepseek Coder v2 scores higher in 0 categories and Gemma 4 31B IT in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Gemma 4 31B IT leads 60.5 to 38.2.

Side by side

Deepseek Coder v2Gemma 4 31B IT
ProviderDeepSeekGoogle
Noometry Index35.943.5
Released2024-06-172026-04-02
WeightsOpenOpen
Context window—262K
Max output—33K
Input $ / M tokens—$0.09
Output $ / M tokens—$0.34
Results tracked2435

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

Category by category

Coding Gemma 4 31B IT leads

Deepseek Coder v2: 38.1 (#183), Gemma 4 31B IT: 42.3 (#108)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Coding12511459
LMArena WebDev—1366
SciCode—43.4%
WeirdML—52.3%
BigCodeBench Instruct48.2%—
BigCodeBench Complete59.7%—
ALE-Bench—925.5
HumanEval+82.3%—
MBPP+75.1%—

Reasoning Gemma 4 31B IT leads

Deepseek Coder v2: 23.6 (#176), Gemma 4 31B IT: 27.2 (#122)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Hard Prompts12071448
Kagi LLM Benchmark—63.5%
NYT Connections (extended)—70.6%
CritPt—1.4%
Chess Puzzles—5%
Thematic Generalization—53%
DTBench—82.7%
LMCA—39.3%
Surface Evolver Bench—30.6%
Epoch Capabilities Index—142.74
WinoGrande83.7%—

Math Gemma 4 31B IT leads

Deepseek Coder v2: 34.9 (#190), Gemma 4 31B IT: 43.2 (#81)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Math12411465
OTIS Mock AIME 2024-2025—73.3%
GSM8K94.5%—

Knowledge Gemma 4 31B IT leads

Deepseek Coder v2: 32.3 (#212), Gemma 4 31B IT: 37.9 (#151)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Expert11811465
GPQA Diamond—75.8%
SimpleQA Verified—10.4%
Vectara Hallucination Rate—7.4%
ARC (AI2) Challenge64.3%—

Multimodal Not comparable

Deepseek Coder v2: —, Gemma 4 31B IT: 41.6 (#34)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Vision—1277
LMArena Document—1425

Multilingual Gemma 4 31B IT leads

Deepseek Coder v2: 36.3 (#240), Gemma 4 31B IT: 53.8 (#57)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Non-English11821431
LMArena Chinese12011476
LMArena French11851435
LMArena Russian11881460
LMArena Spanish11531444
LMArena German1164—
LMArena Japanese1126—
LMArena Korean1104—

Instruction Following Gemma 4 31B IT leads

Deepseek Coder v2: 61.7 (#242), Gemma 4 31B IT: 75.5 (#61)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Instruction Following11801433

Long Context Gemma 4 31B IT leads

Deepseek Coder v2: 37.0 (#224), Gemma 4 31B IT: 44.2 (#71)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Longer Query12191446

Writing & Preference Gemma 4 31B IT leads

Deepseek Coder v2: 38.2 (#253), Gemma 4 31B IT: 60.5 (#96)

BenchmarkDeepseek Coder v2Gemma 4 31B IT
LMArena Text11911443
LMArena Creative Writing11201415
LMArena Multi-Turn11771452
EQ-Bench Creative Writing—1368
EQ-Bench 4—1120

Frequently asked questions

Is Deepseek Coder v2 better than Gemma 4 31B IT?

Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 35.9 on the Noometry Index.

Is Deepseek Coder v2 or Gemma 4 31B IT better for coding?

Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 38.1 in the Noometry coding category.

How many benchmarks do Deepseek Coder v2 and Gemma 4 31B IT share?

14 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Gemma 4 31B IT has 35.

Related comparisons

Go deeper