Noometry

Model comparison

DeepSeek-V3.2-Exp vs Gemini 3.1 Pro Preview

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 44.3 on the Noometry Index. DeepSeek-V3.2-Exp costs 16× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Last verified . 42 shared benchmarks.

DeepSeek-V3.2-Exp DeepSeek

44.3

Rank #78 Confirmed

Gemini 3.1 Pro Preview Google

56.7

Rank #23 Confirmed

Summary

  • They share 42 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and Gemini 3.1 Pro Preview in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 22.1.
  • The biggest single-benchmark swing is ARC-AGI-2: 4% for DeepSeek-V3.2-Exp and 77.1% for Gemini 3.1 Pro Preview.
  • DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 164K.
  • DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.2-ExpGemini 3.1 Pro Preview
ProviderDeepSeekGoogle
Noometry Index44.356.7
Released2025-09-292026-02-19
WeightsOpenProprietary
Context window164K1.05M
Max output66K66K
Input $ / M tokens$0.26$2
Output $ / M tokens$0.38$12
Results tracked4971

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

Category by category

Coding DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 46.5 (#65), Gemini 3.1 Pro Preview: 42.5 (#99)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
LMArena WebDev13621447
SciCode38.9%58.9%
WeirdML39.5%72.1%
LMArena Coding14541484
SWE-bench Verified—75.6%
DeepSWE—11.7%
SWE-bench Verified (bash only)70%—
Aider Polyglot74.2%—
SWE-bench Multilingual59%—
GSO—22.6%
MirrorCode—8.9%
ALE-Bench—1,161
AlgoTune—2.02

Agentic & Tool Use Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 32.7 (#59), Gemini 3.1 Pro Preview: 37.7 (#34)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
Terminal-Bench39.6%80.2%
APEX-Agents21.3%35.3%
Vending-Bench 21,0343,774
Berkeley Function Calling Leaderboard56.7%—
TheAgentCompany42.9%—
τ²-bench Banking—26%
DeepResearch Bench—47.8%
PostTrainBench—22%
BALROG—57%
ExploitBench—26.1%
GBAEval—0.8%
GDP.pdf—17%
LMArena Search—1211
METR Time Horizons—77%

Reasoning Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 22.1 (#208), Gemini 3.1 Pro Preview: 71.7 (#12)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
ARC-AGI-24%77.1%
NYT Connections (extended)36.7%97.4%
ARC-AGI-157%98%
CritPt2.9%17.7%
Chess Puzzles14%55%
Thematic Generalization65%79.4%
LMArena Hard Prompts14341485
DTBench87.7%97.1%
LMCA29.1%53.8%
Epoch Capabilities Index146.27154.77
SimpleBench—79.6%
Kagi LLM Benchmark52.2%—
EnigmaEval—36.8%
EBR-Bench—14.3%
Mystery Game Puzzles—34%
ForecastBench—59

Math Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 41.7 (#87), Gemini 3.1 Pro Preview: 62.1 (#34)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
MathArena Final-Answer Competitions57.7%86.5%
OTIS Mock AIME 2024-202587.8%95.6%
ProofBench8%26%
LMArena Math14351485
FrontierMath (Feb 2025 set)22.1%36.9%
FrontierMath Tier 4 (v1)2.1%16.7%
FrontierMath (Tiers 1-3)—59.6%
FrontierMath Tier 4—26.8%

Knowledge Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 51.7 (#66), Gemini 3.1 Pro Preview: 71.8 (#3)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
GPQA Diamond83.4%94.4%
Vectara Hallucination Rate5.3%10.4%
LMArena Expert14361485
Humanity's Last Exam—46.4%
SimpleQA Verified—73.5%

Multimodal Not comparable

DeepSeek-V3.2-Exp: —, Gemini 3.1 Pro Preview: 37.9 (#69)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
LMArena Vision—1296
Blueprint-Bench 2—26.5%
Furniture Assembly—26.7%
LMArena Document—1444

Multilingual Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 52.2 (#90), Gemini 3.1 Pro Preview: 57.0 (#12)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
LMArena Non-English14091477
LMArena Chinese14611529
LMArena French14331487
LMArena German14401491
LMArena Japanese13741493
LMArena Korean13711455
LMArena Russian14241498
LMArena Spanish14401479

Instruction Following Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 74.5 (#93), Gemini 3.1 Pro Preview: 77.0 (#32)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
LMArena Instruction Following14131466

Long Context Too close to call

DeepSeek-V3.2-Exp: 47.6 (#16), Gemini 3.1 Pro Preview: 47.4 (#18)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
CL-bench13.2%20.8%
CL-bench Life9.5%16.9%
LMArena Longer Query14281483
Fiction.LiveBench83.3%—

Writing & Preference Gemini 3.1 Pro Preview leads

DeepSeek-V3.2-Exp: 62.4 (#77), Gemini 3.1 Pro Preview: 66.1 (#37)

BenchmarkDeepSeek-V3.2-ExpGemini 3.1 Pro Preview
LMArena Text14251481
LMArena Creative Writing14031482
EQ-Bench Creative Writing15151491
LMArena Multi-Turn14271488
EQ-Bench 4—1142

Frequently asked questions

Is DeepSeek-V3.2-Exp better than Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 44.3 on the Noometry Index. DeepSeek-V3.2-Exp costs 16× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Which is cheaper, DeepSeek-V3.2-Exp or Gemini 3.1 Pro Preview?

DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

Is DeepSeek-V3.2-Exp or Gemini 3.1 Pro Preview better for coding?

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 42.5 in the Noometry coding category.

Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 164K.

How many benchmarks do DeepSeek-V3.2-Exp and Gemini 3.1 Pro Preview share?

42 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Gemini 3.1 Pro Preview has 71.

Related comparisons

Go deeper