Noometry

Model comparison

DeepSeek V4 Pro vs Gemini 2.5 Pro

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 45.0 on the Noometry Index.

Last verified . 38 shared benchmarks.

DeepSeek V4 Pro DeepSeek

54.3

Rank #31 Confirmed

Gemini 2.5 Pro Google

45.0

Rank #75 Confirmed

Summary

  • They share 38 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 7 categories and Gemini 2.5 Pro in 2 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where DeepSeek V4 Pro leads 64.8 to 32.5.
  • The biggest single-benchmark swing is ARC-AGI-2: 61.3% for DeepSeek V4 Pro and 4.9% for Gemini 2.5 Pro.
  • DeepSeek V4 Pro is cheaper at $0.66 / $1.98 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.
  • Gemini 2.5 Pro accepts more context: 1.05M tokens versus 1M.
  • DeepSeek V4 Pro has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4 ProGemini 2.5 Pro
ProviderDeepSeekGoogle
Noometry Index54.345.0
Released2026-04-242025-03-25
WeightsOpenProprietary
Context window1M1.05M
Max output393K66K
Input $ / M tokens$0.66$1.25
Output $ / M tokens$1.98$10
Results tracked4878

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

Category by category

Coding DeepSeek V4 Pro leads

DeepSeek V4 Pro: 52.4 (#34), Gemini 2.5 Pro: 42.4 (#101)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
SWE-bench Verified77.6%57.6%
LMArena WebDev15821227
SciCode51%42.8%
WeirdML66.2%54%
LMArena Coding14701452
ALE-Bench1,403785.52
FrontierCode28.6%—
SWE-bench Verified (bash only)—53.6%
Aider Polyglot—83.1%
GSO—3.9%
LiveBench Coding—85.9%
CadEval—64%
AlgoTune—1.51

Agentic & Tool Use DeepSeek V4 Pro leads

DeepSeek V4 Pro: 32.8 (#58), Gemini 2.5 Pro: 29.2 (#88)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
Vending-Bench 23,285573.64
Terminal-Bench—32.6%
APEX-Agents47.3%—
GDPval—23.3%
Remote Labor Index—0.8%
TheAgentCompany—30.3%
τ²-bench Banking—13.7%
DeepResearch Bench—42.8%
BALROG—43.3%
LMArena Search—1142
METR Time Horizons—55.4%

Reasoning DeepSeek V4 Pro leads

DeepSeek V4 Pro: 56.5 (#24), Gemini 2.5 Pro: 28.8 (#99)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
ARC-AGI-261.3%4.9%
Kagi LLM Benchmark53.5%70.3%
ARC-AGI-190.5%41%
CritPt18%2%
Chess Puzzles47%20%
LMArena Hard Prompts14611455
DTBench93.9%82.4%
LMCA45.5%34.8%
Epoch Capabilities Index155.31145.32
ForecastBench56.161.3
SimpleBench—62.4%
NYT Connections (extended)91.3%—
EnigmaEval—5.6%
LiveBench Reasoning—89.8%
Mystery Game Puzzles43%—
LiveBench Data Analysis—79.9%
Surface Evolver Bench40%—
LiveBench—82.3%

Math DeepSeek V4 Pro leads

DeepSeek V4 Pro: 64.8 (#30), Gemini 2.5 Pro: 32.5 (#213)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
FrontierMath (Tiers 1-3)64.6%24.6%
FrontierMath Tier 426.8%0%
OTIS Mock AIME 2024-202598.6%84.7%
LMArena Math14551450
MathArena Final-Answer Competitions76.6%—
ProofBench50%—
Omni-MATH—41.6%
LiveBench Math—90.2%
MATH Level 5—95.9%
FrontierMath (Feb 2025 set)—14.1%
FrontierMath Tier 4 (v1)—4.2%

Knowledge DeepSeek V4 Pro leads

DeepSeek V4 Pro: 59.5 (#31), Gemini 2.5 Pro: 56.0 (#46)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
GPQA Diamond91.7%85.3%
Vectara Hallucination Rate8.6%7%
LMArena Expert14641452
Humanity's Last Exam—21.6%
SimpleQA Verified52.9%—
MMLU-Pro—86.3%
Confabulations—10.6%
GPQA (HELM)—74.9%

Multimodal Not comparable

DeepSeek V4 Pro: —, Gemini 2.5 Pro: 45.2 (#18)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
LMArena Vision—1263
GeoBench—86%
VPCT—48%
LMArena Document—1421
SpatialViz-Bench—44.7%

Multilingual Too close to call

DeepSeek V4 Pro: 54.4 (#45), Gemini 2.5 Pro: 55.3 (#31)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
LMArena Non-English14391451
LMArena Chinese14861507
LMArena French14721472
LMArena German14581487
LMArena Japanese14451461
LMArena Korean14471434
LMArena Russian14531461
LMArena Spanish14581473

Instruction Following DeepSeek V4 Pro leads

DeepSeek V4 Pro: 76.1 (#47), Gemini 2.5 Pro: 75.0 (#75)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
LMArena Instruction Following14481437
LiveBench Instruction Following—80.6%
IFEval—84%

Long Context Gemini 2.5 Pro leads

DeepSeek V4 Pro: 45.0 (#51), Gemini 2.5 Pro: 59.8 (#5)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
LMArena Longer Query14581449
Fiction.LiveBench—91.7%
CL-bench Life13.5%—

Writing & Preference DeepSeek V4 Pro leads

DeepSeek V4 Pro: 65.5 (#46), Gemini 2.5 Pro: 63.7 (#62)

BenchmarkDeepSeek V4 ProGemini 2.5 Pro
LMArena Text14511458
LMArena Creative Writing14461454
EQ-Bench Creative Writing15531421
LMArena Multi-Turn14671453
Short-Story Creative Writing—83.8%
WildBench—85.7%
EQ-Bench 41166—
LiveBench Language—67.8%

Frequently asked questions

Is DeepSeek V4 Pro better than Gemini 2.5 Pro?

DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 45.0 on the Noometry Index.

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

DeepSeek V4 Pro is cheaper. It lists at $0.66 per million input tokens and $1.98 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.

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

DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 42.4 in the Noometry coding category.

Which has the bigger context window?

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

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

38 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Gemini 2.5 Pro has 78.

Related comparisons

Go deeper