Model comparison

DeepSeek V4 Flash vs StarCoder 2 3B

DeepSeek V4 Flash has enough public results to be ranked (#35); StarCoder 2 3B does not yet, so treat this comparison as directional.

Last verified . 1 shared benchmarks.

DeepSeek V4 Flash DeepSeek

53.6

Rank #35 Confirmed

StarCoder 2 3B NVIDIA

36.3

Unranked Sparse

Summary

  • They share 1 benchmark with published results for both. DeepSeek V4 Flash scores higher in 1 category and StarCoder 2 3B in 0 categories; one gap is clear of the uncertainty.
  • The widest gap is in coding, where DeepSeek V4 Flash leads 47.9 to 33.5.

Side by side

DeepSeek V4 Flash and StarCoder 2 3B specifications
DeepSeek V4 FlashStarCoder 2 3B
ProviderDeepSeekNVIDIA
Noometry Index53.636.3
Released2026-04-242024-02-22
WeightsOpenOpen
Context window1M—
Max output393K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked417

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), StarCoder 2 3B: 33.5

Coding benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
FrontierCode18.8%—
LMArena WebDev1582—
SciCode49.9%—
WeirdML63%—
LMArena Coding1457—
BigCodeBench Complete—21.4%
ALE-Bench1,306—
HumanEval+—27.4%

Reasoning Not comparable

DeepSeek V4 Flash: 53.7 (#30), StarCoder 2 3B: —

Reasoning benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
Epoch Capabilities Index154.4988.76
ARC-AGI-261.4%—
SimpleBench61.1%—
Kagi LLM Benchmark52.2%—
NYT Connections (extended)89.6%—
ARC-AGI-189%—
CritPt16.6%—
Chess Puzzles33%—
LMArena Hard Prompts1444—
Mystery Game Puzzles34%—
DTBench90.9%—
LMCA41.7%—
WinoGrande—57.1%

Math Not comparable

DeepSeek V4 Flash: 60.3 (#37), StarCoder 2 3B: —

Math benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
FrontierMath (Tiers 1-3)57.5%—
FrontierMath Tier 424.4%—
MathArena Final-Answer Competitions76.5%—
OTIS Mock AIME 2024-202594.4%—
ProofBench56%—
LMArena Math1427—
GSM8K—21.6%

Knowledge Not comparable

DeepSeek V4 Flash: 55.4 (#48), StarCoder 2 3B: —

Knowledge benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
GPQA Diamond91%—
SimpleQA Verified33.6%—
LMArena Expert1441—
ARC (AI2) Challenge—34.2%
MMLU—36.6%

Multilingual Not comparable

DeepSeek V4 Flash: 53.0 (#72), StarCoder 2 3B: —

Multilingual benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
LMArena Non-English1420—
LMArena Chinese1468—
LMArena French1439—
LMArena German1418—
LMArena Japanese1406—
LMArena Korean1384—
LMArena Russian1428—
LMArena Spanish1436—

Instruction Following Not comparable

DeepSeek V4 Flash: 74.9 (#81), StarCoder 2 3B: —

Instruction Following benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
LMArena Instruction Following1421—

Long Context Not comparable

DeepSeek V4 Flash: 43.8 (#85), StarCoder 2 3B: —

Long Context benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
LMArena Longer Query1434—

Writing & Preference Not comparable

DeepSeek V4 Flash: 63.8 (#61), StarCoder 2 3B: —

Writing & Preference benchmarks
BenchmarkDeepSeek V4 FlashStarCoder 2 3B
LMArena Text1432—
LMArena Creative Writing1403—
EQ-Bench Creative Writing1559—
LMArena Multi-Turn1449—

Frequently asked questions

Is DeepSeek V4 Flash better than StarCoder 2 3B?

DeepSeek V4 Flash has enough public results to be ranked (#35); StarCoder 2 3B does not yet, so treat this comparison as directional.

Is DeepSeek V4 Flash or StarCoder 2 3B better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 33.5 in the Noometry coding category.

How many benchmarks do DeepSeek V4 Flash and StarCoder 2 3B share?

1 benchmark has published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and StarCoder 2 3B has 7.

Related comparisons

Go deeper