Noometry

Model comparison

DeepSeek-R1 vs Gemini 3 Flash Preview

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 42.3 on the Noometry Index.

Last verified . 29 shared benchmarks.

DeepSeek-R1 DeepSeek

42.3

Rank #115 Confirmed

Gemini 3 Flash Preview Google

52.3

Rank #40 Confirmed

Summary

  • They share 29 benchmarks with published results for both. DeepSeek-R1 scores higher in 1 category and Gemini 3 Flash Preview in 8 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3 Flash Preview leads 49.2 to 18.6.
  • The biggest single-benchmark swing is ARC-AGI-1: 21.2% for DeepSeek-R1 and 84.7% for Gemini 3 Flash Preview.
  • DeepSeek-R1 is cheaper at $0.50 / $2.15 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.

Side by side

DeepSeek-R1Gemini 3 Flash Preview
ProviderDeepSeekGoogle
Noometry Index42.352.3
Released2025-01-202025-12-17
WeightsProprietaryProprietary
Context window164K1.05M
Max output64K66K
Input $ / M tokens$0.50$0.50
Output $ / M tokens$2.15$3
Results tracked5259

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

Category by category

Coding Gemini 3 Flash Preview leads

DeepSeek-R1: 46.3 (#68), Gemini 3 Flash Preview: 50.9 (#42)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
WeirdML41.6%61.6%
LMArena Coding14271460
ALE-Bench804.121,367
SWE-bench Verified—75.4%
SWE-bench Verified (bash only)—75.8%
Aider Polyglot71.4%—
LMArena WebDev—1439
SWE-bench Multilingual—72.7%
SciCode35.7%—
GSO—9.8%
LiveBench Coding66.7%—
AlgoTune1.7—

Agentic & Tool Use Gemini 3 Flash Preview leads

DeepSeek-R1: 30.7 (#75), Gemini 3 Flash Preview: 38.7 (#29)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
DeepResearch Bench35.1%49.8%
BALROG34.9%48.1%
Terminal-Bench—64.3%
τ²-bench Airline—82.5%
τ²-bench Banking—27.3%
τ²-bench Retail—76.8%
τ²-bench Telecom—91.2%
GDP.pdf—10%
LMArena Search—1198
METR Time Horizons53.8%—
Vending-Bench 2—3,635

Reasoning Gemini 3 Flash Preview leads

DeepSeek-R1: 18.6 (#278), Gemini 3 Flash Preview: 49.2 (#37)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
ARC-AGI-21.3%33.6%
SimpleBench40.8%61.1%
ARC-AGI-121.2%84.7%
LMArena Hard Prompts14161465
Epoch Capabilities Index141.29151.8
ForecastBench6058.5
Kagi LLM Benchmark69.4%—
NYT Connections (extended)—83.1%
CritPt1.1%—
Chess Puzzles—40%
LiveBench Reasoning83.2%—
Mystery Game Puzzles—26%
DTBench—89.1%
LiveBench Data Analysis69.8%—
LMCA—43.1%
LiveBench71.6%—

Math Gemini 3 Flash Preview leads

DeepSeek-R1: 43.8 (#79), Gemini 3 Flash Preview: 51.7 (#55)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
OTIS Mock AIME 2024-202566.4%95.6%
LMArena Math14001473
FrontierMath (Tiers 1-3)—51.2%
FrontierMath Tier 4—17.1%
MathArena Final-Answer Competitions—67.6%
ProofBench—15%
Omni-MATH42.4%—
LiveBench Math80.7%—
MATH Level 596.6%—
FrontierMath (Feb 2025 set)—35.6%
FrontierMath Tier 4 (v1)—4.2%

Knowledge Gemini 3 Flash Preview leads

DeepSeek-R1: 44.5 (#87), Gemini 3 Flash Preview: 58.8 (#33)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
GPQA Diamond76.3%89.4%
Vectara Hallucination Rate11.3%13.5%
LMArena Expert13941462
SimpleQA Verified—66.8%
MMLU-Pro79.3%—
Confabulations12.7%—
GPQA (HELM)66.6%—

Multimodal Not comparable

DeepSeek-R1: —, Gemini 3 Flash Preview: 45.5 (#16)

BenchmarkDeepSeek-R1Gemini 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-R1: 52.4 (#85), Gemini 3 Flash Preview: 55.7 (#27)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
LMArena Non-English14121458
LMArena Chinese14421511
LMArena French14171477
LMArena German14041497
LMArena Japanese13911489
LMArena Korean13601443
LMArena Russian14231480
LMArena Spanish14111469

Instruction Following Gemini 3 Flash Preview leads

DeepSeek-R1: 72.0 (#143), Gemini 3 Flash Preview: 75.7 (#56)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
LMArena Instruction Following13821437
LiveBench Instruction Following80.5%—
IFEval78.4%—

Long Context DeepSeek-R1 leads

DeepSeek-R1: 45.4 (#36), Gemini 3 Flash Preview: 44.4 (#67)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
LMArena Longer Query13911452
Fiction.LiveBench75%—

Writing & Preference Gemini 3 Flash Preview leads

DeepSeek-R1: 61.4 (#88), Gemini 3 Flash Preview: 65.5 (#45)

BenchmarkDeepSeek-R1Gemini 3 Flash Preview
LMArena Text14281466
LMArena Creative Writing14051457
LMArena Multi-Turn14051471
Short-Story Creative Writing83%—
EQ-Bench Creative Writing1500—
WildBench82.8%—
LiveBench Language48.5%—

Frequently asked questions

Is DeepSeek-R1 better than Gemini 3 Flash Preview?

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 42.3 on the Noometry Index.

Which is cheaper, DeepSeek-R1 or Gemini 3 Flash Preview?

DeepSeek-R1 is cheaper. It lists at $0.50 per million input tokens and $2.15 per million output tokens; Gemini 3 Flash Preview lists at $0.50 and $3.

Is DeepSeek-R1 or Gemini 3 Flash Preview better for coding?

Gemini 3 Flash Preview scores higher on coding benchmarks: 50.9 versus 46.3 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-R1 and Gemini 3 Flash Preview share?

29 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and Gemini 3 Flash Preview has 59.

Related comparisons

Go deeper