Enter opens the highlighted result. Esc closes.

Gemini 3.7 Flash vs Gemini 4 Carbon

Gemini 4 Carbon leads Gemini 3.7 Flash on coding, 86.6 to 53.9 on the Vibe Coding Index, and Gemini 3.7 Flash costs 2.7x less per output token ($3.75 vs $10.00 per million). Updated November 14, 2026; 53 benchmarks have a result for both.

Replace Gemini 3.7 Flash Replace Gemini 4 Carbon
Gemini 3.7 FlashGoogle
Coding
53.9
Overall
66.3
Output $/M
$3.75
1
Gemini 4 CarbonGoogle
Coding
86.6
Overall
88.3
Output $/M
$10.00
52
vs boards won of 53 shared

Gemini 3.7 Flash or Gemini 4 Carbon: which to pick

Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 53.9 for Gemini 3.7 Flash, though on 4 benchmarks against 5.

Pick Gemini 3.7 Flash for output tokens 2.7x cheaper ($3.75 vs $10.00 per million) and work measured by Long context reasoning (AA-LCR), where it leads.

Model AGemini 3.7 FlashGoogle53.9/1005 independent benchmarks$1.78 per solved taskModel BGemini 4 CarbonGoogle86.6/1004 independent benchmarksNo cost per solved task
Shared results
53
Gemini 3.7 Flash only
60
Gemini 4 Carbon only
20
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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 3.7 Flash 53.9 from 5 benchmarks, Gemini 4 Carbon 86.6 from 4 benchmarks.
Widest category
Math, 29.6 points apart on average over 5 shared benchmarks.
Widest single gap
Terminal-Bench 4 (Vals AI): Gemini 3.7 Flash 12.1% against Gemini 4 Carbon 63.1%, 68.0 points apart on a 0 to 100 scale.
Output price
Gemini 3.7 Flash $3.75 against Gemini 4 Carbon $10.00 per million output tokens, 2.7x apart.
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 3.7 FlashBarsGemini 4 CarbonTies
Coding010
Intelligence01
Agentic work05
Reasoning08
Math05
Knowledge05
Web design05
General113
All shared boards152
CodingLike-for-likeSix of these boards feed the Vibe Coding Index010
Coding benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
LiveBench codingNot measured91.0%Only Gemini 4 Carbon measured
LMArena WebDev15881768Gemini 4 Carbon
AA Coding Index76.1Not measuredOnly Gemini 3.7 Flash measured
LiveCodeBench (Vals AI)88.7%Not measuredOnly Gemini 3.7 Flash measured
SWE-bench Verified (Vals AI)80.8%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench 4 (Vals AI)12.1%63.1%Gemini 4 Carbon
Vibe Code Bench (Vals AI)70.4%89.1%Gemini 4 Carbon
Code migration (Vals AI)34.8%Not measuredOnly Gemini 3.7 Flash measured
ProgramBench (Vals AI)0.0%Not measuredOnly Gemini 3.7 Flash measured
IOI (Vals AI)67.8%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench 2.1 (Vals AI)77.5%Not measuredOnly Gemini 3.7 Flash measured
SkillsBench (Vals AI)65.9%Not measuredOnly Gemini 3.7 Flash measured
CyberBench (Vals AI)43.8%Not measuredOnly Gemini 3.7 Flash measured
SciCode59.8%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench 4.0 (Artificial Analysis run)13.6%65.2%Gemini 4 Carbon
FrontierCode 2.0Not measured40.6%Only Gemini 4 Carbon measured
FrontierCode43.6%Not measuredOnly Gemini 3.7 Flash measured
FrontierSWE20.3%61.2%Gemini 4 Carbon
MirrorCode41.1%Not measuredOnly Gemini 3.7 Flash measured
ALE-Bench904.32374.7Gemini 4 Carbon
LMArena Text: coding15001564Gemini 4 Carbon
LMArena Text: software and IT15001552Gemini 4 Carbon
LMArena WebDev: React15831792Gemini 4 Carbon
LMArena WebDev: HTML15931770Gemini 4 Carbon
LMArena WebDev: from an image1600Not measuredOnly Gemini 3.7 Flash measured
SciCode13.8%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench 4.011.2%Not measuredOnly Gemini 3.7 Flash measured
CursorBench 4.0Not measured60.4%Only Gemini 4 Carbon measured
ProgramBench0.0%Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 23 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
AA Intelligence IndexNot measured60.0Only Gemini 4 Carbon measured
Epoch Capabilities Index158.4169.8Gemini 4 Carbon
Agentic workLike-for-likePublic boards, shown as published, kept out of the index05
Agentic work benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
AA Agentic Index35.2Not measuredOnly Gemini 3.7 Flash measured
Design Arena: agents building web apps1201Not measuredOnly Gemini 3.7 Flash measured
Design Arena: agents building full-stack apps1185Not measuredOnly Gemini 3.7 Flash measured
Design Arena: agents building mobile apps12261284Gemini 4 Carbon
Design Arena: agents building Android apps1233Not measuredOnly Gemini 3.7 Flash measured
Design Arena: agents building Godot games1236Not measuredOnly Gemini 3.7 Flash measured
Design Arena: agents building games1177Not measuredOnly Gemini 3.7 Flash measured
Design Arena: PowerPoint slides1069Not measuredOnly Gemini 3.7 Flash measured
Design Arena: spreadsheets10141224Gemini 4 Carbon
LiveBench researchNot measured92.5%Only Gemini 4 Carbon measured
LiveBench browserNot measured80.0%Only Gemini 4 Carbon measured
LiveBench long agenticNot measured72.7%Only Gemini 4 Carbon measured
SREBench (Vals AI)4.6%Not measuredOnly Gemini 3.7 Flash measured
Finance Agent (Vals AI)59.0%Not measuredOnly Gemini 3.7 Flash measured
Tax Agent Bench (Vals AI)22.8%51.8%Gemini 4 Carbon
Legal Research Bench (Vals AI)34.6%55.8%Gemini 4 Carbon
Harvey's Legal Agent Benchmark (Vals AI)Not measured17.5%Only Gemini 4 Carbon measured
Excel Modeling Benchmark (Vals AI)71.3%Not measuredOnly Gemini 3.7 Flash measured
Terminal-Bench-Science (Vals AI)7.1%Not measuredOnly Gemini 3.7 Flash measured
APEX-Agents67.8%84.2%Gemini 4 Carbon
Remote Labor Index5.0%Not measuredOnly Gemini 3.7 Flash measured
Vending-Bench 2Not measured$16,484Only Gemini 4 Carbon measured
Neither model has a result on the other 22 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index08
Reasoning benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
GPQA Diamond93.9%Not measuredOnly Gemini 3.7 Flash measured
LiveBench reasoningNot measured90.6%Only Gemini 4 Carbon measured
MMMU Pro (Vals AI)89.0%Not measuredOnly Gemini 3.7 Flash measured
GPQA Diamond (Artificial Analysis run)94.5%Not measuredOnly Gemini 3.7 Flash measured
Humanity's Last Exam (Artificial Analysis run)47.9%Not measuredOnly Gemini 3.7 Flash measured
HLE-Diamond (Artificial Analysis run)Not measured69.1%Only Gemini 4 Carbon measured
CritPt v2, hard subset (Artificial Analysis run)Not measured69.0%Only Gemini 4 Carbon measured
GPQA Diamond (Epoch AI run)94.8%96.5%Gemini 4 Carbon
ARC-AGI95.5%98.5%Gemini 4 Carbon
ARC-AGI-284.6%94.2%Gemini 4 Carbon
SimpleBenchNot measured92.6%Only Gemini 4 Carbon measured
Chess puzzles66.0%78.0%Gemini 4 Carbon
Mystery game puzzles37.0%73.0%Gemini 4 Carbon
EBR-benchNot measured87.6%Only Gemini 4 Carbon measured
LMArena Text: hard prompts14981551Gemini 4 Carbon
LMArena Text: expert15151582Gemini 4 Carbon
LMArena Text: hard prompts in English14881545Gemini 4 Carbon
LMArena Vision: diagrams1329Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: homework1325Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 7 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index05
Math benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
OTIS Mock AIME97.2%100.0%Gemini 4 Carbon
FrontierMath, tiers 1 to 371.6%90.5%Gemini 4 Carbon
FrontierMath, tier 436.6%95.1%Gemini 4 Carbon
ProofBench58.0%Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: math1506Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: mathematical work15001538Gemini 4 Carbon
MathArena23.9%85.4%Gemini 4 Carbon
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
Knowledge benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
MMLU Pro90.1%Not measuredOnly Gemini 3.7 Flash measured
LegalBench (Vals AI)87.3%Not measuredOnly Gemini 3.7 Flash measured
TaxEval (Vals AI)74.7%Not measuredOnly Gemini 3.7 Flash measured
MortgageTax (Vals AI)66.7%Not measuredOnly Gemini 3.7 Flash measured
MedCode (Vals AI)53.4%Not measuredOnly Gemini 3.7 Flash measured
AA-OmniscienceNot measured53.5Only Gemini 4 Carbon measured
SimpleQA Verified69.2%77.7%Gemini 4 Carbon
LMArena Text: science15111535Gemini 4 Carbon
LMArena Text: medicine and healthcare14991525Gemini 4 Carbon
LMArena Text: legal and government14981542Gemini 4 Carbon
LMArena Text: business and finance14811520Gemini 4 Carbon
FACTS Benchmark SuiteNot measured78.4%Only Gemini 4 Carbon measured
PRBench LegalNot measured53.3%Only Gemini 4 Carbon measured
Neither model has a result on the other 8 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index05
Web design benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
Design Arena: websites12951319Gemini 4 Carbon
Design Arena: UI components12871357Gemini 4 Carbon
Design Arena: game dev13121326Gemini 4 Carbon
Design Arena: data viz1308Not measuredOnly Gemini 3.7 Flash measured
Design Arena: 3D12871434Gemini 4 Carbon
Design Arena: all code categories12981348Gemini 4 Carbon
Design Arena: SVG1278Not measuredOnly Gemini 3.7 Flash measured
Design Arena: ASCII art1267Not measuredOnly Gemini 3.7 Flash measured
LiveBench designNot measured80.2%Only Gemini 4 Carbon measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index113
General benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
Vals IndexNot measured70.5%Only Gemini 4 Carbon measured
LiveBench data analysisNot measured86.5%Only Gemini 4 Carbon measured
MedScribe (Vals AI)83.9%Not measuredOnly Gemini 3.7 Flash measured
Long context reasoning (AA-LCR)83.0%79.3%Gemini 3.7 Flash
GDPval-AA14041733Gemini 4 Carbon
DTBench96.8%Not measuredOnly Gemini 3.7 Flash measured
LMCA50.4%Not measuredOnly Gemini 3.7 Flash measured
Furniture assembly26.7%77.5%Gemini 4 Carbon
LMArena Text14891538Gemini 4 Carbon
LMArena Text: instruction following14851539Gemini 4 Carbon
LMArena Text: creative writing14851533Gemini 4 Carbon
LMArena Vision1313Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: English14841530Gemini 4 Carbon
LMArena Text: non-English14861534Gemini 4 Carbon
LMArena Text: Chinese15531613Gemini 4 Carbon
LMArena Text: German1514Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Japanese1515Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Korean1483Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: French1494Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Spanish1484Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: Russian15131560Gemini 4 Carbon
LMArena Text: Polish1507Not measuredOnly Gemini 3.7 Flash measured
LMArena Text: multi-turn14871569Gemini 4 Carbon
LMArena Text: longer queries14941549Gemini 4 Carbon
LMArena Text: writing and language14891542Gemini 4 Carbon
LMArena Text: entertainment and media14741527Gemini 4 Carbon
LMArena Vision: English1309Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: Chinese1388Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: reading text1321Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: humor1300Not measuredOnly Gemini 3.7 Flash measured
LMArena Vision: creative writing1320Not measuredOnly Gemini 3.7 Flash measured
LMArena Document1476Not measuredOnly Gemini 3.7 Flash measured
Neither model has a result on the other 17 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 3.7 Flash and Gemini 4 Carbon
BenchmarkGemini 3.7 FlashGemini 4 CarbonLeader
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 3.7 Flash$0.0026Fits in one request
Gemini 4 Carbon$0.0070Fits in one request

Gemini 3.7 Flash costs less

Repository review

50K input + 3K output tokens

Gemini 3.7 Flash$0.049Fits in one request
Gemini 4 Carbon$0.130Fits in one request

Gemini 3.7 Flash costs less

Agent loop

200K input + 10K output tokens

Gemini 3.7 Flash$0.188Fits in one request
Gemini 4 Carbon$0.500Fits in one request

Gemini 3.7 Flash costs less

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 3.7 Flash and Gemini 4 Carbon
SpecificationGoogle Gemini 3.7 FlashGoogle Gemini 4 Carbon
MakerGoogleGoogle
WeightsNo public weights listedNo public weights listed
ReleasedAugust 13, 2026November 10, 2026
Context window1.05M1.05M
Input $/Mtok$0.75$2.00
Output $/Mtok$3.75$10.00
Coding index53.986.6
Independent measurements behind it54
Cost per solved task$1.78Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 3.7 Flash better than Gemini 4 Carbon for coding?

Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 53.9 for Gemini 3.7 Flash, though on 4 benchmarks against 5. Pick Gemini 3.7 Flash for output tokens 2.7x cheaper ($3.75 vs $10.00 per million) and work measured by Long context reasoning (AA-LCR), where it leads.

Which is cheaper, Gemini 3.7 Flash or Gemini 4 Carbon?

Gemini 3.7 Flash costs $0.75 per million input tokens and $3.75 per million output tokens; Gemini 4 Carbon costs $2.00 and $10.00. Cost per solved task is known for Gemini 3.7 Flash only.

Which has the bigger context window, Gemini 3.7 Flash or Gemini 4 Carbon?

Neither: both take 1.05M tokens.

How many benchmarks is the Gemini 3.7 Flash vs Gemini 4 Carbon comparison based on?

53 benchmarks have a result for both models as of November 14, 2026. Gemini 3.7 Flash's coding index rests on 5 independent benchmarks and Gemini 4 Carbon's on 4. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/gemini-3-7-flash-vs-gemini-4-carbon" title="Gemini 3.7 Flash vs Gemini 4 Carbon, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/gemini-3-7-flash-vs-gemini-4-carbon">LLMBenchmark.ai, Gemini 3.7 Flash vs Gemini 4 Carbon</a></p>