Noometry

Model comparison

DeepSeek V4 Flash vs Gemini 3.1 Flash Lite

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 40.8 on the Noometry Index.

Last verified . 31 shared benchmarks.

DeepSeek V4 Flash DeepSeek

53.6

Rank #35 Confirmed

Gemini 3.1 Flash Lite Google

40.8

Rank #144 Confirmed

Summary

  • They share 31 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 8 categories and Gemini 3.1 Flash Lite in 0 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 22.9.
  • The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4 Flash and 8.2% for Gemini 3.1 Flash Lite.
  • DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.25 / $1.50 for Gemini 3.1 Flash Lite.
  • Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 1M.
  • DeepSeek V4 Flash has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4 FlashGemini 3.1 Flash Lite
ProviderDeepSeekGoogle
Noometry Index53.640.8
Released2026-04-242026-03-03
WeightsOpenProprietary
Context window1M1.05M
Max output393K66K
Input $ / M tokens$0.15$0.25
Output $ / M tokens$0.60$1.50
Results tracked4138

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), Gemini 3.1 Flash Lite: 37.8 (#188)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
LMArena WebDev15821256
SciCode49.9%41.9%
WeirdML63%52.2%
LMArena Coding14571400
ALE-Bench1,306797.73
FrontierCode18.8%—

Agentic & Tool Use Not comparable

DeepSeek V4 Flash: —, Gemini 3.1 Flash Lite: 30.2 (#79)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
DeepResearch Bench—37.3%

Reasoning DeepSeek V4 Flash leads

DeepSeek V4 Flash: 53.7 (#30), Gemini 3.1 Flash Lite: 22.9 (#186)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
Kagi LLM Benchmark52.2%67.2%
NYT Connections (extended)89.6%8.2%
CritPt16.6%1.1%
Chess Puzzles33%25%
LMArena Hard Prompts14441407
DTBench90.9%76.8%
LMCA41.7%35%
Epoch Capabilities Index154.49144.47
ARC-AGI-261.4%—
SimpleBench61.1%—
ARC-AGI-189%—
EnigmaEval—3%
Thematic Generalization—63.3%
Mystery Game Puzzles34%—
ForecastBench—54.4

Math DeepSeek V4 Flash leads

DeepSeek V4 Flash: 60.3 (#37), Gemini 3.1 Flash Lite: 40.7 (#90)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
FrontierMath (Tiers 1-3)57.5%27.7%
OTIS Mock AIME 2024-202594.4%80%
LMArena Math14271428
FrontierMath Tier 424.4%—
MathArena Final-Answer Competitions76.5%—
ProofBench56%—

Knowledge DeepSeek V4 Flash leads

DeepSeek V4 Flash: 55.4 (#48), Gemini 3.1 Flash Lite: 41.9 (#104)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
GPQA Diamond91%81.8%
LMArena Expert14411398
Humanity's Last Exam—8.6%
SimpleQA Verified33.6%—
Vectara Hallucination Rate—8.2%

Multimodal Not comparable

DeepSeek V4 Flash: —, Gemini 3.1 Flash Lite: 39.4 (#60)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
LMArena Vision—1240

Multilingual Too close to call

DeepSeek V4 Flash: 53.0 (#72), Gemini 3.1 Flash Lite: 52.3 (#86)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
LMArena Non-English14201411
LMArena Chinese14681461
LMArena French14391424
LMArena German14181429
LMArena Japanese14061413
LMArena Korean13841392
LMArena Russian14281420
LMArena Spanish14361421

Instruction Following DeepSeek V4 Flash leads

DeepSeek V4 Flash: 74.9 (#81), Gemini 3.1 Flash Lite: 72.7 (#131)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
LMArena Instruction Following14211377

Long Context DeepSeek V4 Flash leads

DeepSeek V4 Flash: 43.8 (#85), Gemini 3.1 Flash Lite: 42.5 (#122)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
LMArena Longer Query14341394

Writing & Preference DeepSeek V4 Flash leads

DeepSeek V4 Flash: 63.8 (#61), Gemini 3.1 Flash Lite: 60.9 (#94)

BenchmarkDeepSeek V4 FlashGemini 3.1 Flash Lite
LMArena Text14321416
LMArena Creative Writing14031401
LMArena Multi-Turn14491417
EQ-Bench Creative Writing1559—

Frequently asked questions

Is DeepSeek V4 Flash better than Gemini 3.1 Flash Lite?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 40.8 on the Noometry Index.

Which is cheaper, DeepSeek V4 Flash or Gemini 3.1 Flash Lite?

DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemini 3.1 Flash Lite lists at $0.25 and $1.50.

Is DeepSeek V4 Flash or Gemini 3.1 Flash Lite better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 37.8 in the Noometry coding category.

Which has the bigger context window?

Gemini 3.1 Flash Lite does, with 1.05M tokens against 1M.

How many benchmarks do DeepSeek V4 Flash and Gemini 3.1 Flash Lite share?

31 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Gemini 3.1 Flash Lite has 38.

Related comparisons

Go deeper