Noometry

Model comparison

DeepSeek-V3 vs Gemini 3 Flash Preview

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 39.5 on the Noometry Index. DeepSeek-V3 costs 2.8× less per token, which makes it the better buy when Gemini 3 Flash Preview's lead doesn't matter for your workload.

Last verified . 27 shared benchmarks.

DeepSeek-V3 DeepSeek

39.5

Rank #166 Confirmed

Gemini 3 Flash Preview Google

52.3

Rank #40 Confirmed

Summary

  • They share 27 benchmarks with published results for both. DeepSeek-V3 scores higher in 0 categories and Gemini 3 Flash Preview in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3 Flash Preview leads 49.2 to 20.5.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for DeepSeek-V3 and 95.6% for Gemini 3 Flash Preview.
  • DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $0.50 / $3 for Gemini 3 Flash Preview.
  • Gemini 3 Flash Preview accepts more context: 1.05M tokens versus 164K.
  • DeepSeek-V3 has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3Gemini 3 Flash Preview
ProviderDeepSeekGoogle
Noometry Index39.552.3
Released2024-12-262025-12-17
WeightsOpenProprietary
Context window164K1.05M
Max output164K66K
Input $ / M tokens$0.24$0.50
Output $ / M tokens$0.90$3
Results tracked6059

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

Category by category

Coding Gemini 3 Flash Preview leads

DeepSeek-V3: 42.3 (#106), Gemini 3 Flash Preview: 50.9 (#42)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
WeirdML36.1%61.6%
LMArena Coding13681460
SWE-bench Verified—75.4%
SWE-bench Verified (bash only)—75.8%
Aider Polyglot55.1%—
LMArena WebDev—1439
SWE-bench Multilingual—72.7%
SciCode35.8%—
GSO—9.8%
BigCodeBench Instruct50%—
LiveBench Coding70.9%—
BigCodeBench Complete62.2%—
ALE-Bench—1,367
HumanEval+86.6%—
MBPP+73%—

Agentic & Tool Use Not comparable

DeepSeek-V3: —, Gemini 3 Flash Preview: 38.7 (#29)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
Terminal-Bench—64.3%
τ²-bench Airline—82.5%
τ²-bench Banking—27.3%
τ²-bench Retail—76.8%
τ²-bench Telecom—91.2%
DeepResearch Bench—49.8%
BALROG—48.1%
GDP.pdf—10%
LMArena Search—1198
METR Time Horizons49.6%—
Vending-Bench 2—3,635

Reasoning Gemini 3 Flash Preview leads

DeepSeek-V3: 20.5 (#236), Gemini 3 Flash Preview: 49.2 (#37)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
SimpleBench27.2%61.1%
LMArena Hard Prompts13651465
DTBench64.8%89.1%
LMCA15.5%43.1%
Epoch Capabilities Index135.94151.8
ForecastBench59.158.5
ARC-AGI-2—33.6%
Kagi LLM Benchmark52.3%—
NYT Connections (extended)—83.1%
ARC-AGI-1—84.7%
CritPt0%—
Chess Puzzles—40%
LiveBench Reasoning65.8%—
Mystery Game Puzzles—26%
LiveBench Data Analysis60.9%—
BIG-Bench Hard87.5%—
HellaSwag88.9%—
LiveBench66.9%—
PIQA84.7%—
WinoGrande85.2%—

Math Gemini 3 Flash Preview leads

DeepSeek-V3: 32.1 (#219), Gemini 3 Flash Preview: 51.7 (#55)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
OTIS Mock AIME 2024-202537.8%95.6%
LMArena Math13731473
FrontierMath (Feb 2025 set)1.7%35.6%
FrontierMath (Tiers 1-3)—51.2%
FrontierMath Tier 4—17.1%
MathArena Final-Answer Competitions—67.6%
ProofBench—15%
Omni-MATH40.3%—
LiveBench Math73.5%—
MATH Level 575.5%—
FrontierMath Tier 4 (v1)—4.2%

Knowledge Gemini 3 Flash Preview leads

DeepSeek-V3: 37.5 (#155), Gemini 3 Flash Preview: 58.8 (#33)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
GPQA Diamond67.6%89.4%
Vectara Hallucination Rate6.1%13.5%
LMArena Expert13511462
SimpleQA Verified—66.8%
MMLU-Pro72.3%—
Confabulations26.1%—
GPQA (HELM)53.8%—
ARC (AI2) Challenge95.3%—
MMLU87.2%—
TriviaQA82.9%—

Multimodal Not comparable

DeepSeek-V3: —, Gemini 3 Flash Preview: 45.5 (#16)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
LMArena Vision—1285
GeoBench—88%
VPCT—72.6%
Blueprint-Bench 2—0%
LMArena Document—1413

Multilingual Gemini 3 Flash Preview leads

DeepSeek-V3: 48.5 (#143), Gemini 3 Flash Preview: 55.7 (#27)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
LMArena Non-English13581458
LMArena Chinese13911511
LMArena French13851477
LMArena German13741497
LMArena Japanese13331489
LMArena Korean13191443
LMArena Russian13731480
LMArena Spanish13581469

Instruction Following Gemini 3 Flash Preview leads

DeepSeek-V3: 72.8 (#130), Gemini 3 Flash Preview: 75.7 (#56)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
LMArena Instruction Following13451437
LiveBench Instruction Following81.5%—
IFEval83.2%—

Long Context Gemini 3 Flash Preview leads

DeepSeek-V3: 34.0 (#253), Gemini 3 Flash Preview: 44.4 (#67)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
LMArena Longer Query13521452
Fiction.LiveBench50%—

Writing & Preference Gemini 3 Flash Preview leads

DeepSeek-V3: 57.4 (#130), Gemini 3 Flash Preview: 65.5 (#45)

BenchmarkDeepSeek-V3Gemini 3 Flash Preview
LMArena Text13751466
LMArena Creative Writing13641457
LMArena Multi-Turn13891471
Short-Story Creative Writing77%—
EQ-Bench Creative Writing1472—
WildBench83%—
LiveBench Language49.1%—

Frequently asked questions

Is DeepSeek-V3 better than Gemini 3 Flash Preview?

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 39.5 on the Noometry Index. DeepSeek-V3 costs 2.8× less per token, which makes it the better buy when Gemini 3 Flash Preview's lead doesn't matter for your workload.

Which is cheaper, DeepSeek-V3 or Gemini 3 Flash Preview?

DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; Gemini 3 Flash Preview lists at $0.50 and $3.

Is DeepSeek-V3 or Gemini 3 Flash Preview better for coding?

Gemini 3 Flash Preview scores higher on coding benchmarks: 50.9 versus 42.3 in the Noometry coding category.

Which has the bigger context window?

Gemini 3 Flash Preview does, with 1.05M tokens against 164K.

How many benchmarks do DeepSeek-V3 and Gemini 3 Flash Preview share?

27 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Gemini 3 Flash Preview has 59.

Related comparisons

Go deeper