Model comparison

Gemini 3.5 Flash vs Gemini 3 Flash Preview

Gemini 3.5 Flash is the stronger model overall, scoring 54.2 to 52.3 on the Noometry Index. Gemini 3 Flash Preview costs 3.0× less per token, which makes it the better buy when Gemini 3.5 Flash's lead doesn't matter for your workload.

Last verified . 45 shared benchmarks.

Gemini 3.5 Flash Google

54.2

Rank #32 Confirmed

Gemini 3 Flash Preview Google

52.3

Rank #40 Confirmed

Summary

  • They share 45 benchmarks with published results for both. Gemini 3.5 Flash scores higher in 7 categories and Gemini 3 Flash Preview in 3 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in agentic & tool use, where Gemini 3 Flash Preview leads 38.7 to 24.7.
  • The biggest single-benchmark swing is ARC-AGI-2: 72.1% for Gemini 3.5 Flash and 33.6% for Gemini 3 Flash Preview.
  • Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $1.50 / $9 for Gemini 3.5 Flash.

Side by side

Gemini 3.5 Flash and Gemini 3 Flash Preview specifications
Gemini 3.5 FlashGemini 3 Flash Preview
ProviderGoogleGoogle
Noometry Index54.252.3
Released2026-05-192025-12-17
WeightsProprietaryProprietary
Context window1.05M1.05M
Max output66K66K
Input $ / M tokens$1.50$0.50
Output $ / M tokens$9$3
Results tracked5459

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

Category by category

Coding Gemini 3 Flash Preview leads

Gemini 3.5 Flash: 49.4 (#49), Gemini 3 Flash Preview: 50.9 (#42)

Coding benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
SWE-bench Verified79.3%75.4%
LMArena WebDev14991439
WeirdML62.6%61.6%
LMArena Coding14921460
ALE-Bench911.021,367
DeepSWE37.4%—
SWE-bench Verified (bash only)—75.8%
SWE-bench Multilingual—72.7%
SciCode53.1%—
GSO—9.8%

Agentic & Tool Use Gemini 3 Flash Preview leads

Gemini 3.5 Flash: 24.7 (#114), Gemini 3 Flash Preview: 38.7 (#29)

Agentic & Tool Use benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
GDP.pdf14%10%
Vending-Bench 25,3963,635
Terminal-Bench—64.3%
APEX-Agents27.5%—
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%
GBAEval6.7%—
LMArena Search—1198

Reasoning Gemini 3.5 Flash leads

Gemini 3.5 Flash: 62.8 (#18), Gemini 3 Flash Preview: 49.2 (#37)

Reasoning benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
ARC-AGI-272.1%33.6%
SimpleBench76.7%61.1%
NYT Connections (extended)92.6%83.1%
ARC-AGI-192.5%84.7%
Chess Puzzles50%40%
LMArena Hard Prompts14881465
Mystery Game Puzzles32%26%
DTBench94.7%89.1%
LMCA47.1%43.1%
Epoch Capabilities Index154.46151.8
ForecastBench5958.5
CritPt13.1%—
EnigmaEval25.4%—
EBR-Bench4.8%—
Surface Evolver Bench58.1%—

Math Gemini 3.5 Flash leads

Gemini 3.5 Flash: 60.7 (#36), Gemini 3 Flash Preview: 51.7 (#55)

Math benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
FrontierMath (Tiers 1-3)62.8%51.2%
FrontierMath Tier 426.8%17.1%
MathArena Final-Answer Competitions76.3%67.6%
OTIS Mock AIME 2024-202595.6%95.6%
ProofBench31%15%
LMArena Math15041473
FrontierMath (Feb 2025 set)39%35.6%
FrontierMath Tier 4 (v1)14.6%4.2%

Knowledge Gemini 3.5 Flash leads

Gemini 3.5 Flash: 66.3 (#11), Gemini 3 Flash Preview: 58.8 (#33)

Knowledge benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
GPQA Diamond92.8%89.4%
SimpleQA Verified66.2%66.8%
LMArena Expert14951462
Vectara Hallucination Rate—13.5%

Multimodal Too close to call

Gemini 3.5 Flash: 45.7 (#15), Gemini 3 Flash Preview: 45.5 (#16)

Multimodal benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
LMArena Vision13101285
Blueprint-Bench 233.6%0%
LMArena Document14631413
GeoBench—88%
VPCT—72.6%

Multilingual Gemini 3.5 Flash leads

Gemini 3.5 Flash: 57.0 (#13), Gemini 3 Flash Preview: 55.7 (#27)

Multilingual benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
LMArena Non-English14761458
LMArena Chinese15261511
LMArena French14901477
LMArena German14921497
LMArena Japanese14861489
LMArena Korean14511443
LMArena Russian14931480
LMArena Spanish14801469

Instruction Following Gemini 3.5 Flash leads

Gemini 3.5 Flash: 77.0 (#30), Gemini 3 Flash Preview: 75.7 (#56)

Instruction Following benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
LMArena Instruction Following14671437

Long Context Gemini 3.5 Flash leads

Gemini 3.5 Flash: 45.4 (#38), Gemini 3 Flash Preview: 44.4 (#67)

Long Context benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
LMArena Longer Query14821452

Writing & Preference Too close to call

Gemini 3.5 Flash: 65.5 (#47), Gemini 3 Flash Preview: 65.5 (#45)

Writing & Preference benchmarks
BenchmarkGemini 3.5 FlashGemini 3 Flash Preview
LMArena Text14821466
LMArena Creative Writing14701457
LMArena Multi-Turn14811471
EQ-Bench 41087—

Frequently asked questions

Is Gemini 3.5 Flash better than Gemini 3 Flash Preview?

Gemini 3.5 Flash is the stronger model overall, scoring 54.2 to 52.3 on the Noometry Index. Gemini 3 Flash Preview costs 3.0× less per token, which makes it the better buy when Gemini 3.5 Flash's lead doesn't matter for your workload.

Which is cheaper, Gemini 3.5 Flash or Gemini 3 Flash Preview?

Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Gemini 3.5 Flash lists at $1.50 and $9.

Is Gemini 3.5 Flash or Gemini 3 Flash Preview better for coding?

Gemini 3 Flash Preview scores higher on coding benchmarks: 50.9 versus 49.4 in the Noometry coding category.

Which has the bigger context window?

Both accept 1.05M tokens.

How many benchmarks do Gemini 3.5 Flash and Gemini 3 Flash Preview share?

45 benchmarks have published results for both models. Gemini 3.5 Flash has 54 scored results on Noometry and Gemini 3 Flash Preview has 59.

Related comparisons

Go deeper