LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Gemini 2.5 Flash vs Gemini 3.5 Flash Lite

Updated October 21, 2026. 48 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Gemini 2.5 Flash Replace Gemini 3.5 Flash Lite
Gemini 2.5 FlashGoogle
Coding
28.7
Overall
15.5
Output $/M
$2.50
1
Gemini 3.5 Flash LiteGoogle
Coding
33.6
Overall
22.2
Output $/M
$2.50
47
vs boards won of 48 shared

Decision reading

Pick Gemini 3.5 Flash Lite for coding: it scores 33.6 on the Vibe Coding Index against 28.7 for Gemini 2.5 Flash.

Pick Gemini 2.5 Flash when the benchmark that matters is MedScribe (Vals AI), where it leads.

Model AGemini 2.5 FlashGoogle28.7/1001 independent benchmark$0.466 per solved taskModel BGemini 3.5 Flash LiteGoogle33.6/1001 independent benchmark$0.41 per solved task
Shared results
48
Gemini 2.5 Flash only
21
Gemini 3.5 Flash Lite only
35
Like-for-like categories
6 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Gemini 2.5 Flash 28.7 from 1 benchmark, Gemini 3.5 Flash Lite 33.6 from 1 benchmark.
Widest category
Reasoning, 6.8 points apart on average over 9 shared benchmarks.
Widest single gap
ARC-AGI: Gemini 2.5 Flash 33.3% against Gemini 3.5 Flash Lite 53.5%, 20.2 points apart on a 0 to 100 scale.
Output price
Both cost $2.50 per million output tokens.
Context window
Both take 1.05M tokens.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryGemini 2.5 FlashBarsGemini 3.5 Flash LiteTies
Coding02
Intelligence02
Reasoning09
Math02
Knowledge05
General127
All shared boards147
CodingLike-for-likeFour of these boards feed the Vibe Coding Index02
Coding benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
SWE-bench Verified28.7%Not measuredOnly Gemini 2.5 Flash measured
LMArena WebDevNot measured1435Only Gemini 3.5 Flash Lite measured
AA Coding IndexNot measured49.3Only Gemini 3.5 Flash Lite measured
LiveCodeBench (Vals AI)Not measured79.0%Only Gemini 3.5 Flash Lite measured
SWE-bench Verified (Vals AI)Not measured75.0%Only Gemini 3.5 Flash Lite measured
Vibe Code Bench (Vals AI)Not measured37.2%Only Gemini 3.5 Flash Lite measured
Code migration (Vals AI)Not measured6.2%Only Gemini 3.5 Flash Lite measured
ProgramBench (Vals AI)Not measured0.0%Only Gemini 3.5 Flash Lite measured
Terminal-Bench 2.1 (Vals AI)Not measured50.2%Only Gemini 3.5 Flash Lite measured
CyberBench (Vals AI)Not measured59.1%Only Gemini 3.5 Flash Lite measured
LiveCodeBench (Artificial Analysis run)69.5%Not measuredOnly Gemini 2.5 Flash measured
SciCodeNot measured41.3%Only Gemini 3.5 Flash Lite measured
Terminal-Bench Hard (Artificial Analysis run)13.6%Not measuredOnly Gemini 2.5 Flash measured
Terminal-Bench17.1%Not measuredOnly Gemini 2.5 Flash measured
WeirdMLNot measured39.0%Only Gemini 3.5 Flash Lite measured
ALE-BenchNot measured765.3Only Gemini 3.5 Flash Lite measured
LMArena Text: coding14231453Gemini 3.5 Flash Lite
LMArena Text: software and IT14251451Gemini 3.5 Flash Lite
LMArena WebDev: ReactNot measured1408Only Gemini 3.5 Flash Lite measured
LMArena WebDev: from an imageNot measured1448Only Gemini 3.5 Flash Lite measured
Neither model has a result on the other 18 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
Intelligence benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
AA Intelligence Index15.522.2Gemini 3.5 Flash Lite
Epoch Capabilities Index142.9144.9Gemini 3.5 Flash Lite
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Agentic work benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
AA Agentic IndexNot measured14.3Only Gemini 3.5 Flash Lite measured
Finance Agent (Vals AI)Not measured47.4%Only Gemini 3.5 Flash Lite measured
Tax Agent Bench (Vals AI)Not measured7.2%Only Gemini 3.5 Flash Lite measured
Legal Research Bench (Vals AI)Not measured13.9%Only Gemini 3.5 Flash Lite measured
Excel Modeling Benchmark (Vals AI)Not measured43.2%Only Gemini 3.5 Flash Lite measured
τ²-Bench Telecom31.6%Not measuredOnly Gemini 2.5 Flash measured
APEX-AgentsNot measured29.3%Only Gemini 3.5 Flash Lite measured
BALROG33.5%Not measuredOnly Gemini 2.5 Flash measured
Vending-Bench 2$549Not measuredOnly Gemini 2.5 Flash measured
Neither model has a result on the other 30 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index09
Reasoning benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
GPQA DiamondNot measured83.8%Only Gemini 3.5 Flash Lite measured
MMMU Pro (Vals AI)Not measured83.6%Only Gemini 3.5 Flash Lite measured
GPQA Diamond (Artificial Analysis run)79.0%83.8%Gemini 3.5 Flash Lite
Humanity's Last Exam (Artificial Analysis run)12.1%18.8%Gemini 3.5 Flash Lite
GPQA Diamond (Epoch AI run)Not measured83.3%Only Gemini 3.5 Flash Lite measured
ARC-AGI33.3%53.5%Gemini 3.5 Flash Lite
ARC-AGI-22.5%10.3%Gemini 3.5 Flash Lite
SimpleBench41.2%Not measuredOnly Gemini 2.5 Flash measured
Chess puzzlesNot measured21.0%Only Gemini 3.5 Flash Lite measured
Mystery game puzzlesNot measured19.0%Only Gemini 3.5 Flash Lite measured
LMArena Text: hard prompts14221441Gemini 3.5 Flash Lite
LMArena Text: expert14261434Gemini 3.5 Flash Lite
LMArena Text: hard prompts in English14201440Gemini 3.5 Flash Lite
LMArena Vision: diagrams12451284Gemini 3.5 Flash Lite
LMArena Vision: homework12441291Gemini 3.5 Flash Lite
Neither model has a result on the other 6 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index02
Math benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
AA Math Index73.3Not measuredOnly Gemini 2.5 Flash measured
AIME 2025 (Artificial Analysis run)73.3%Not measuredOnly Gemini 2.5 Flash measured
OTIS Mock AIMENot measured71.1%Only Gemini 3.5 Flash Lite measured
FrontierMath, tiers 1 to 3Not measured26.0%Only Gemini 3.5 Flash Lite measured
FrontierMath, tier 4Not measured0.0%Only Gemini 3.5 Flash Lite measured
ProofBenchNot measured13.0%Only Gemini 3.5 Flash Lite measured
LMArena Text: math14111434Gemini 3.5 Flash Lite
LMArena Text: mathematical work14221433Gemini 3.5 Flash Lite
Neither model has a result on the other 5 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
Knowledge benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
MMLU ProNot measured85.8%Only Gemini 3.5 Flash Lite measured
LegalBench (Vals AI)Not measured84.1%Only Gemini 3.5 Flash Lite measured
CorpFin (Vals AI)Not measured60.7%Only Gemini 3.5 Flash Lite measured
TaxEval (Vals AI)Not measured72.6%Only Gemini 3.5 Flash Lite measured
MortgageTax (Vals AI)Not measured68.7%Only Gemini 3.5 Flash Lite measured
MedCode (Vals AI)40.4%43.5%Gemini 3.5 Flash Lite
MMLU Pro (Artificial Analysis run)83.2%Not measuredOnly Gemini 2.5 Flash measured
LMArena Text: science14331451Gemini 3.5 Flash Lite
LMArena Text: medicine and healthcare14281438Gemini 3.5 Flash Lite
LMArena Text: legal and government14301431Gemini 3.5 Flash Lite
LMArena Text: business and finance14041423Gemini 3.5 Flash Lite
Neither model has a result on the other 6 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Web design benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
Design Arena: websites1074Not measuredOnly Gemini 2.5 Flash measured
Design Arena: UI components1005Not measuredOnly Gemini 2.5 Flash measured
Design Arena: game dev1005Not measuredOnly Gemini 2.5 Flash measured
Design Arena: data viz1060Not measuredOnly Gemini 2.5 Flash measured
Design Arena: 3D1021Not measuredOnly Gemini 2.5 Flash measured
Design Arena: all code categories1059Not measuredOnly Gemini 2.5 Flash measured
Design Arena: SVG1018Not measuredOnly Gemini 2.5 Flash measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index127
General benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
Public Benefits Bench (Vals AI)Not measured51.8%Only Gemini 3.5 Flash Lite measured
SAGE (Vals AI)44.8%47.3%Gemini 3.5 Flash Lite
MedScribe (Vals AI)83.0%70.9%Gemini 2.5 Flash
IFBench50.3%Not measuredOnly Gemini 2.5 Flash measured
Long context reasoning (AA-LCR)65.3%76.0%Gemini 3.5 Flash Lite
DTBench76.5%83.5%Gemini 3.5 Flash Lite
LMCA27.5%37.6%Gemini 3.5 Flash Lite
LMArena Text14171435Gemini 3.5 Flash Lite
LMArena Text: instruction following14051419Gemini 3.5 Flash Lite
LMArena Text: creative writing14001421Gemini 3.5 Flash Lite
LMArena Vision12351269Gemini 3.5 Flash Lite
LMArena Text: English14181438Gemini 3.5 Flash Lite
LMArena Text: non-English14091427Gemini 3.5 Flash Lite
LMArena Text: Chinese14491468Gemini 3.5 Flash Lite
LMArena Text: German14171454Gemini 3.5 Flash Lite
LMArena Text: Japanese13961426Gemini 3.5 Flash Lite
LMArena Text: Korean13851407Gemini 3.5 Flash Lite
LMArena Text: French14321451Gemini 3.5 Flash Lite
LMArena Text: Spanish14221441Gemini 3.5 Flash Lite
LMArena Text: Russian14151443Gemini 3.5 Flash Lite
LMArena Text: Polish14131437Gemini 3.5 Flash Lite
LMArena Text: multi-turn14081445Gemini 3.5 Flash Lite
LMArena Text: longer queries14191435Gemini 3.5 Flash Lite
LMArena Text: writing and language14051428Gemini 3.5 Flash Lite
LMArena Text: entertainment and media13931414Gemini 3.5 Flash Lite
LMArena Vision: English12391269Gemini 3.5 Flash Lite
LMArena Vision: Chinese12841311Gemini 3.5 Flash Lite
LMArena Vision: reading text12371280Gemini 3.5 Flash Lite
LMArena Vision: humor12251287Gemini 3.5 Flash Lite
LMArena Vision: captioning1277Not measuredOnly Gemini 2.5 Flash measured
LMArena Vision: recognizing things1252Not measuredOnly Gemini 2.5 Flash measured
LMArena Vision: creative writing12411260Gemini 3.5 Flash Lite
Neither model has a result on the other 14 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 2.5 Flash and Gemini 3.5 Flash Lite
BenchmarkGemini 2.5 FlashGemini 3.5 Flash LiteLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Gemini 2.5 Flash$0.0015Fits in one request
Gemini 3.5 Flash Lite$0.0015Fits in one request

Same cost

Repository review

50K input + 3K output tokens

Gemini 2.5 Flash$0.022Fits in one request
Gemini 3.5 Flash Lite$0.022Fits in one request

Same cost

Agent loop

200K input + 10K output tokens

Gemini 2.5 Flash$0.085Fits in one request
Gemini 3.5 Flash Lite$0.085Fits in one request

Same cost

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Gemini 2.5 Flash and Gemini 3.5 Flash Lite
SpecificationGoogle Gemini 2.5 FlashGoogle Gemini 3.5 Flash Lite
MakerGoogleGoogle
WeightsNo public weights listedNo public weights listed
ReleasedJune 17, 2025July 21, 2026
Context window1.05M1.05M
Input $/Mtok$0.30$0.30
Output $/Mtok$2.50$2.50
Coding index28.733.6
Independent measurements behind it11
Cost per solved task$0.466$0.41
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 2.5 Flash better than Gemini 3.5 Flash Lite for coding?

Pick Gemini 3.5 Flash Lite for coding: it scores 33.6 on the Vibe Coding Index against 28.7 for Gemini 2.5 Flash. Pick Gemini 2.5 Flash when the benchmark that matters is MedScribe (Vals AI), where it leads.

Which is cheaper, Gemini 2.5 Flash or Gemini 3.5 Flash Lite?

Gemini 2.5 Flash costs $0.30 per million input tokens and $2.50 per million output tokens; Gemini 3.5 Flash Lite costs $0.30 and $2.50. Per solved SWE-bench Verified task, Gemini 2.5 Flash costs $0.466 and Gemini 3.5 Flash Lite $0.41.

Which has the bigger context window, Gemini 2.5 Flash or Gemini 3.5 Flash Lite?

Neither: both take 1.05M tokens.

How many benchmarks is the Gemini 2.5 Flash vs Gemini 3.5 Flash Lite comparison based on?

48 benchmarks have a result for both models as of October 21, 2026. Gemini 2.5 Flash's coding index rests on 1 independent benchmark and Gemini 3.5 Flash Lite's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next