Noometry

Model comparison

Gemini 2.5 Pro vs Gemini 3.8 Flash

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 45.0 on the Noometry Index.

Last verified . 37 shared benchmarks.

Gemini 2.5 Pro Google

45.0

Rank #75 Confirmed

Gemini 3.8 Flash Google

61.8

Rank #11 Confirmed

Summary

  • They share 37 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 2 categories and Gemini 3.8 Flash in 8 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 28.8.
  • The biggest single-benchmark swing is ARC-AGI-2: 4.9% for Gemini 2.5 Pro and 89.2% for Gemini 3.8 Flash.
  • Gemini 3.8 Flash is cheaper at $0.75 / $3.75 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.

Side by side

Gemini 2.5 ProGemini 3.8 Flash
ProviderGoogleGoogle
Noometry Index45.061.8
Released2025-03-252026-09-02
WeightsProprietaryProprietary
Context window1.05M1.05M
Max output66K66K
Input $ / M tokens$1.25$0.75
Output $ / M tokens$10$3.75
Results tracked7850

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

Category by category

Coding Gemini 3.8 Flash leads

Gemini 2.5 Pro: 42.4 (#101), Gemini 3.8 Flash: 59.2 (#15)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
LMArena WebDev12271584
SciCode42.8%56.6%
WeirdML54%84.8%
LMArena Coding14521510
ALE-Bench785.521,270
SWE-bench Verified57.6%—
DeepSWE—73.8%
FrontierCode—41.2%
SWE-bench Verified (bash only)53.6%—
Aider Polyglot83.1%—
CursorBench—39.6%
FrontierSWE—19.6%
GSO3.9%—
LiveBench Coding85.9%—
CadEval64%—
AlgoTune1.51—

Agentic & Tool Use Gemini 3.8 Flash leads

Gemini 2.5 Pro: 29.2 (#88), Gemini 3.8 Flash: 41.8 (#21)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
Remote Labor Index0.8%5.8%
Vending-Bench 2573.645,094
Terminal-Bench32.6%—
APEX-Agents—64.3%
GDPval23.3%—
TheAgentCompany30.3%—
τ²-bench Banking13.7%—
DeepResearch Bench42.8%—
BALROG43.3%—
GDP.pdf—23.4%
LMArena Search1142—
METR Time Horizons55.4%—

Reasoning Gemini 3.8 Flash leads

Gemini 2.5 Pro: 28.8 (#99), Gemini 3.8 Flash: 76.9 (#5)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
ARC-AGI-24.9%89.2%
ARC-AGI-141%98.5%
CritPt2%18.3%
Chess Puzzles20%61%
LMArena Hard Prompts14551508
DTBench82.4%95.7%
LMCA34.8%52.9%
Epoch Capabilities Index145.32156.71
SimpleBench62.4%—
Kagi LLM Benchmark70.3%—
NYT Connections (extended)—97.4%
EnigmaEval5.6%—
LiveBench Reasoning89.8%—
Mystery Game Puzzles—47%
LiveBench Data Analysis79.9%—
Surface Evolver Bench—76.9%
ForecastBench61.3—
LiveBench82.3%—

Math Gemini 3.8 Flash leads

Gemini 2.5 Pro: 32.5 (#213), Gemini 3.8 Flash: 65.3 (#28)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
FrontierMath (Tiers 1-3)24.6%68.4%
FrontierMath Tier 40%22%
OTIS Mock AIME 2024-202584.7%98.9%
LMArena Math14501528
ProofBench—48%
Omni-MATH41.6%—
LiveBench Math90.2%—
MATH Level 595.9%—
FrontierMath (Feb 2025 set)14.1%—
FrontierMath Tier 4 (v1)4.2%—

Knowledge Gemini 3.8 Flash leads

Gemini 2.5 Pro: 56.0 (#46), Gemini 3.8 Flash: 74.8 (#2)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
GPQA Diamond85.3%95.4%
Humanity's Last Exam21.6%44.5%
LMArena Expert14521524
SimpleQA Verified—69.7%
MMLU-Pro86.3%—
Confabulations10.6%—
Vectara Hallucination Rate7%—
GPQA (HELM)74.9%—

Multimodal Gemini 2.5 Pro leads

Gemini 2.5 Pro: 45.2 (#18), Gemini 3.8 Flash: 40.7 (#45)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
LMArena Vision12631314
GeoBench86%—
VPCT48%—
Blueprint-Bench 2—38.6%
Furniture Assembly—31.7%
LMArena Document1421—
SpatialViz-Bench44.7%—

Multilingual Gemini 3.8 Flash leads

Gemini 2.5 Pro: 55.3 (#31), Gemini 3.8 Flash: 58.0 (#5)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
LMArena Non-English14511491
LMArena Chinese15071554
LMArena French14721498
LMArena German14871493
LMArena Japanese14611502
LMArena Korean14341459
LMArena Russian14611515
LMArena Spanish14731485

Instruction Following Gemini 3.8 Flash leads

Gemini 2.5 Pro: 75.0 (#75), Gemini 3.8 Flash: 78.0 (#13)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
LMArena Instruction Following14371490
LiveBench Instruction Following80.6%—
IFEval84%—

Long Context Gemini 2.5 Pro leads

Gemini 2.5 Pro: 59.8 (#5), Gemini 3.8 Flash: 46.3 (#24)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
LMArena Longer Query14491508
Fiction.LiveBench91.7%—

Writing & Preference Gemini 3.8 Flash leads

Gemini 2.5 Pro: 63.7 (#62), Gemini 3.8 Flash: 72.2 (#15)

BenchmarkGemini 2.5 ProGemini 3.8 Flash
LMArena Text14581499
LMArena Creative Writing14541492
EQ-Bench Creative Writing14211748
LMArena Multi-Turn14531501
Short-Story Creative Writing83.8%—
WildBench85.7%—
LiveBench Language67.8%—

Frequently asked questions

Is Gemini 2.5 Pro better than Gemini 3.8 Flash?

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 45.0 on the Noometry Index.

Which is cheaper, Gemini 2.5 Pro or Gemini 3.8 Flash?

Gemini 3.8 Flash is cheaper. It lists at $0.75 per million input tokens and $3.75 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.

Is Gemini 2.5 Pro or Gemini 3.8 Flash better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 42.4 in the Noometry coding category.

Which has the bigger context window?

Both accept 1.05M tokens.

How many benchmarks do Gemini 2.5 Pro and Gemini 3.8 Flash share?

37 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and Gemini 3.8 Flash has 50.

Related comparisons

Go deeper