Noometry

Model comparison

DeepSeek V4 Flash vs Gemini 2.5 Flash

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 39.3 on the Noometry Index.

Last verified . 29 shared benchmarks.

DeepSeek V4 Flash DeepSeek

53.6

Rank #35 Confirmed

Gemini 2.5 Flash Google

39.3

Rank #170 Confirmed

Summary

  • They share 29 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 6 categories and Gemini 2.5 Flash in 2 categories; 6 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 18.1.
  • The biggest single-benchmark swing is ARC-AGI-2: 61.4% for DeepSeek V4 Flash and 2.5% for Gemini 2.5 Flash.
  • DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.30 / $2.50 for Gemini 2.5 Flash.
  • Gemini 2.5 Flash 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 2.5 Flash
ProviderDeepSeekGoogle
Noometry Index53.639.3
Released2026-04-242025-04-17
WeightsOpenProprietary
Context window1M1.05M
Max output393K66K
Input $ / M tokens$0.15$0.30
Output $ / M tokens$0.60$2.50
Results tracked4154

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 2.5 Flash: 35.8 (#220)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
WeirdML63%41.9%
LMArena Coding14571424
ALE-Bench1,306661.88
FrontierCode18.8%—
SWE-bench Verified (bash only)—28.7%
Aider Polyglot—55.1%
LMArena WebDev1582—
SciCode49.9%—

Agentic & Tool Use Not comparable

DeepSeek V4 Flash: —, Gemini 2.5 Flash: 30.8 (#74)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
Terminal-Bench—17.1%
Berkeley Function Calling Leaderboard—56.2%
TheAgentCompany—41.1%
BALROG—33.5%
Vending-Bench 2—548.84

Reasoning DeepSeek V4 Flash leads

DeepSeek V4 Flash: 53.7 (#30), Gemini 2.5 Flash: 18.1 (#286)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
ARC-AGI-261.4%2.5%
SimpleBench61.1%41.2%
Kagi LLM Benchmark52.2%56.8%
ARC-AGI-189%33.3%
CritPt16.6%1.1%
LMArena Hard Prompts14441422
DTBench90.9%76.5%
LMCA41.7%27.5%
Epoch Capabilities Index154.49143.03
NYT Connections (extended)89.6%—
Chess Puzzles33%—
EnigmaEval—2.7%
Mystery Game Puzzles34%—
ForecastBench—60.6

Math DeepSeek V4 Flash leads

DeepSeek V4 Flash: 60.3 (#37), Gemini 2.5 Flash: 39.9 (#98)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
OTIS Mock AIME 2024-202594.4%73.1%
LMArena Math14271415
FrontierMath (Tiers 1-3)57.5%—
FrontierMath Tier 424.4%—
MathArena Final-Answer Competitions76.5%—
ProofBench56%—
Omni-MATH—38.5%
FrontierMath (Feb 2025 set)—4.8%
FrontierMath Tier 4 (v1)—4.2%

Knowledge DeepSeek V4 Flash leads

DeepSeek V4 Flash: 55.4 (#48), Gemini 2.5 Flash: 36.4 (#168)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
LMArena Expert14411426
GPQA Diamond91%—
Humanity's Last Exam—12.1%
SimpleQA Verified33.6%—
MMLU-Pro—63.9%
Confabulations—16.8%
Vectara Hallucination Rate—7.8%
GPQA (HELM)—39%

Multimodal Not comparable

DeepSeek V4 Flash: —, Gemini 2.5 Flash: 41.8 (#32)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
LMArena Vision—1253
GeoBench—76%
VPCT—46.2%
SpatialViz-Bench—36.9%

Multilingual Too close to call

DeepSeek V4 Flash: 53.0 (#72), Gemini 2.5 Flash: 52.3 (#88)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
LMArena Non-English14201409
LMArena Chinese14681450
LMArena French14391433
LMArena German14181418
LMArena Japanese14061405
LMArena Korean13841385
LMArena Russian14281415
LMArena Spanish14361421

Instruction Following Too close to call

DeepSeek V4 Flash: 74.9 (#81), Gemini 2.5 Flash: 75.7 (#54)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
LMArena Instruction Following14211405
IFEval—89.8%

Long Context Gemini 2.5 Flash leads

DeepSeek V4 Flash: 43.8 (#85), Gemini 2.5 Flash: 47.5 (#17)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
LMArena Longer Query14341419
Fiction.LiveBench—77.8%

Writing & Preference DeepSeek V4 Flash leads

DeepSeek V4 Flash: 63.8 (#61), Gemini 2.5 Flash: 53.8 (#157)

BenchmarkDeepSeek V4 FlashGemini 2.5 Flash
LMArena Text14321417
LMArena Creative Writing14031400
EQ-Bench Creative Writing15591137
LMArena Multi-Turn14491408
Short-Story Creative Writing—76.5%
WildBench—81.7%

Frequently asked questions

Is DeepSeek V4 Flash better than Gemini 2.5 Flash?

DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 39.3 on the Noometry Index.

Which is cheaper, DeepSeek V4 Flash or Gemini 2.5 Flash?

DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Gemini 2.5 Flash lists at $0.30 and $2.50.

Is DeepSeek V4 Flash or Gemini 2.5 Flash better for coding?

DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 35.8 in the Noometry coding category.

Which has the bigger context window?

Gemini 2.5 Flash does, with 1.05M tokens against 1M.

How many benchmarks do DeepSeek V4 Flash and Gemini 2.5 Flash share?

29 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Gemini 2.5 Flash has 54.

Related comparisons

Go deeper