BenchLeader

Gemini 3 Pro vs Gemini 4 Carbon

Full profiles: Gemini 3 Pro (Google) and Gemini 4 Carbon (Google): every benchmark, price and rank.

Updated · Data from 16 sources

As of 10 Nov 2026, Gemini 4 Carbon is ahead of Gemini 3 Pro on quality (68.5 vs 60.2 on the BenchLeader Index). Gemini 4 Carbon leads in every category we measure both in.

Key facts (as of 10 Nov 2026)

  • Gemini 3 Pro scores 60.2 on the BenchLeader Index and Gemini 4 Carbon scores 68.5 (BenchLeader, 10 Nov 2026).
  • Gemini 4 Carbon costs $8.00 per 1M tokens blended (list price, 10 Nov 2026).
  • Of the 7 benchmarks both models have, Gemini 4 Carbon leads 7 (BenchLeader, 10 Nov 2026).
  • On Arena WebDev, Gemini 3 Pro is rated 1441 and Gemini 4 Carbon 1768 (LMArena, as published by 10 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingGemini 4 Carbon75 vs 58 category score; Arena WebDev 1768 vs 1441High
Agents & tool useGemini 4 Carbon78 vs 61 category scoreHigh
ReasoningGemini 4 Carbon77 vs 66 category score; Arena Hard Prompts 1546 vs 1502High
KnowledgeGemini 4 Carbon74 vs 56 category scoreHigh
Chat quality (human preference)Gemini 4 Carbon73 vs 68 on Arena votesMedium
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Gemini 4 Carbon (high) leads on quality: 68.5 vs 60.2.
  • Gemini 3 Pro is stronger in instruction following, maths, multimodal.
  • Gemini 4 Carbon (high) is stronger in agents & tools, coding, human preference, knowledge, reasoning, long context.

At a glance

SpecGemini 3 ProGemini 4 Carbon
DeveloperGoogleGoogle
Released18 Nov 202528 Oct 2026
Context window (tokens)–1,048,576
Compared atGemini 3 ProGemini 4 Carbon (high)
Open weightsNoNo
Price in / out / cached input per 1M–$4.00 / $20.00 / $0.100

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryGemini 3 ProGemini 4 Carbon (high)Leader
Reasoning65.576.9Gemini 4 Carbon
Coding57.874.5Gemini 4 Carbon
Agents & tools61.378.2Gemini 4 Carbon
Maths54.2–not comparable
Knowledge56.274.4Gemini 4 Carbon
Instruction following60.1–not comparable
Human preference67.772.6Gemini 4 Carbon
Multimodal65.6–not comparable
Long context–64.3not comparable

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsKnowledgeHumanpreferenceGemini 3 Pro, Reasoning: 65.5Gemini 3 Pro, Coding: 57.8Gemini 3 Pro, Agents & tools: 61.3Gemini 3 Pro, Knowledge: 56.2Gemini 3 Pro, Human preference: 67.7Gemini 4 Carbon, Reasoning: 76.9Gemini 4 Carbon, Coding: 74.5Gemini 4 Carbon, Agents & tools: 78.2Gemini 4 Carbon, Knowledge: 74.4Gemini 4 Carbon, Human preference: 72.6

Not drawn, measured for fewer than two of these models: Maths, Instruction following, Long context, Multimodal.

ModelIndex

Where sources disagree

Across the 7 benchmarks both have, Gemini 3 Pro leads 0, Gemini 4 Carbon leads 7. Every overall index that rates both puts Gemini 4 Carbon ahead.

Overall indexGemini 3 ProGemini 4 Carbon (high)Leader
BenchLeader Index60.268.5Gemini 4 Carbon
Arena Text14861526Gemini 4 Carbon

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

MeasureGemini 3 ProGemini 4 Carbon (high)
Blended price per 1M tokens–$8.00
Input / output per 1M– / –$4.00 / $20.00
Cached input per 1M–$0.100
Price per task: chat reply–$0.0076
Price per task: summarize a 30-page report–$0.060
Price per task: code edit–$0.054
Price per task: agentic coding session–$0.320
Price per task: structured extraction–$0.012
Agentic coding session, input mostly cached–$0.144
Cost per benchmark task (AA Index)–$7.76
Output tokens per task (AA Index)–123k
Of which reasoning–72k
Time per task (AA Index)–26 min
Output speed–80 tok/s
Time to first answer–66.8 s
Time to a complete answer–156.1 s

Each model is compared at its best-ranked reasoning-effort setting. Gemini 3 Pro scores 60.2; Gemini 3 Pro (high) scores 60.1. Gemini 4 Carbon (high) scores 68.5 and takes 67 s to its first answer token; Gemini 4 Carbon (medium) scores 64.1 in 23 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

131 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherGemini 3 ProGemini 4 Carbon (high)
BenchLeader IndexBenchLeader60.268.5
Agents & tools scoreBenchLeader61.378.2
Coding scoreBenchLeader57.874.5
Human preference scoreBenchLeader67.772.6
Instruction following scoreBenchLeader60.1–
Knowledge scoreBenchLeader56.274.4
Maths scoreBenchLeader54.2–
Multimodal scoreBenchLeader65.6–
Reasoning scoreBenchLeader65.576.9
Long context scoreBenchLeader–64.3
Blended price $/MBenchLeader–$8.00
Output speedBenchLeader–80 tok/s
Time to first answerBenchLeader–66.8 s
Context windowBenchLeader–1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub92.6%–
Humanity's Last ExamScale AI / CAIS37.5%–
SimpleBenchSimpleBench76.4%–
Arena Hard PromptsArena (LMArena)15021546
GPQA Diamond (AA)Artificial Analysis90.8%†–
Humanity's Last Exam (AA)Artificial Analysis39.7%†–
HLE-Diamond (AA)Artificial Analysis–69.1%
GPQA Diamond (Vals)Vals AI91.7%†–
Kagi LLM BenchmarkKagi LLM Benchmark80.1%–
GPQA Diamond (HELM)HELM Capabilities80.3%–
ARC-AGI-1ARC Prize75.0%–
ARC-AGI-2ARC Prize54.0%–
CritPtArtificial Analysis9.1%†–
CritPt v2 hard subset (AA)Artificial Analysis–69.0%
SciPredictScale AI SEAL25.3%–
CL-benchEpoch AI Benchmarking Hub15.8%–
ForecastBenchForecastBench65.0%–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub72.9%–
SciCodeSciCode–64.2%
WeirdMLWeirdML69.9%–
GSO-BenchGSO-Bench18.6%–
Arena CodingArena (LMArena)15171567
Arena WebDevArena (LMArena)14411768
LiveCodeBenchVals AI86.4%–
SWE-bench (Vals)Vals AI76.4%†–
Show all 131 benchmarks
MetricPublisherGemini 3 ProGemini 4 Carbon (high)
Coding
SWE-Bench ProScale AI SEAL43.3%–
SWE-bench Verified (bash only)SWE-bench74.2%–
SWE-bench Verified (any scaffold)SWE-bench77.4%–
Vibe Code Bench v1.1Vals AI14.3%†89.1%
SWE-Bench Pro (private)Scale AI SEAL43.3%–
ALE-BenchEpoch AI Benchmarking Hub1176.8–
AlgoTuneAlgoTune1.8–
FrontierSWEEpoch AI Benchmarking Hub–61.2%
Terminal-Bench 4.0 (AA)Artificial Analysis–65.2%
Design Arena WebsiteDesign Arena1249–
FrontierCode 2.0 (AA)Artificial Analysis–40.6%
LiveCodeBench (AA)Artificial Analysis91.8%†–
Agents & tools
Terminal-BenchTerminal-Bench69.4%–
GDPvalOpenAI40.3%–
Remote Labor IndexScale AI / CAIS1.3%–
APEX-AgentsMercor–84.2%†
Arena AgentArena (LMArena)–16.6
Terminal-Bench HardArtificial Analysis41.7%†–
τ²-Bench Telecom (AA)Artificial Analysis87.1%†–
MCP AtlasScale AI SEAL70.3%–
BFCL OverallBerkeley Function Calling Leaderboard72.5%–
τ²-benchτ²-bench82.5%†–
τ³-bench Bankingτ²-bench18.0%†–
GDPval-AA v2.1Artificial Analysis–1733
Finance Agent v2.1Vals AI–67.2%
Harvey's Legal Agent Benchmark v1.1Vals AI–17.5%
Legal Research BenchVals AI–55.8%
Poker AgentVals AI1078.9%–
Tax Agent BenchVals AI–51.8%
Terminal-Bench 2.0 (Vals)Vals AI55.1%†–
Terminal-Bench 4.0 (Vals)Vals AI–63.1%
Arena SearchArena (LMArena)1201–
BALROGBALROG58.1%–
DeepResearch BenchDeepResearch Bench46.3%†–
METR Time HorizonsMETR71.0%–
Vending-Bench 2Andon Labs5478.216484.2†
AutomationBenchArtificial Analysis–77.8%
GDP.pdf (AA)Artificial Analysis–28.4%
AA-Briefcase v1.1Artificial Analysis–1611
GDPxGDPx by SemiAnalysis–10.9%
Design Arena AgenticDesign Arena1073–
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–64.8%
FACTS SearchGoogle–94.4%†
Maths
OTIS Mock AIMEEpoch AI Benchmarking Hub91.4%–
ProofBenchVals AI20.0%–
AIME (Vals)Vals AI96.7%†–
MGSMVals AI93.9%†–
AIME 2026MathArena91.7%–
HMMT February 2026MathArena86.4%–
MathArena ApexMathArena23.4%–
Omni-MATH (HELM)HELM Capabilities55.6%–
MATH 500Vals AI96.4%†–
Arena MathsArena (LMArena)1476–
Knowledge
AA-Omniscience v1.1Artificial Analysis–53.5
MMLU-ProVals AI90.1%–
LegalBenchVals AI87.0%†–
CorpFinVals AI63.7%†–
TaxEvalVals AI72.6%†–
MedQAVals AI96.0%†–
PRBench FinanceScale AI SEAL39.2%–
PRBench LegalScale AI SEAL40.6%–
MultiNRCScale AI SEAL59.0%–
MMLU-Pro (HELM)HELM Capabilities90.3%–
CaseLaw v2Vals AI53.1%†–
Excel Modeling Benchmark v1.1Vals AI–76.5%
MedCodeVals AI52.2%†–
MedScribeVals AI72.0%†–
AA-Omniscience v1.1: accuracyArtificial Analysis–58.3%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–88.6%
Global-MMLU-Lite (AA)Artificial Analysis92.2%†–
Instruction following
IFBenchArtificial Analysis70.4%†–
MultiChallengeScale AI SEAL65.7%–
TutorBenchScale AI SEAL53.7%–
IFEval (HELM)HELM Capabilities87.6%–
WildBench (HELM)HELM Capabilities85.9%–
Arena Instruction FollowingArena (LMArena)14731525
Human preference
Arena TextArena (LMArena)14861526
Arena Creative WritingArena (LMArena)14841515
Arena Multi-turnArena (LMArena)14961564
Arena Longer QueriesArena (LMArena)14911537
Multimodal
Arena VisionArena (LMArena)1305–
MMMU-ProArtificial Analysis80.2%†89.3%†
VISTAScale AI SEAL51.5%–
MMMU-Pro (official)MMMU81.0%–
MMMU-Pro (Vals)Vals AI87.5%†–
MortgageTaxVals AI69.1%†–
AudioMCScale AI SEAL54.6%†–
AudioMC (text output)Scale AI SEAL54.6%†–
VTBScale AI SEAL26.9%–
GeoBenchGeoBench84.0%–
VPCTEpoch AI Benchmarking Hub91.0%–
Blueprint-Bench 2Epoch AI Benchmarking Hub–58.6%†
Long context
AA-LCR v1.1Artificial Analysis76.0%†79.3%
Arena DocumentArena (LMArena)1451–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub152.9–
AA Intelligence Index v5.0Artificial Analysis–60.0
AA Intelligence Index v4.3.2Artificial Analysis28.0†–
AA Progress Index v1Artificial Analysis290.7†350.8
Vals IndexVals AI–70.5
HELM Capabilities meanHELM Capabilities79.9%–
Vals Finance IndexVals AI–67.0%
Vals RSI IndexVals AI–27.7%
Safety & honesty
FORTRESSScale AI SEAL41.7%–
MASKScale AI SEAL42.6%–
PropensityBenchScale AI SEAL52.9%–
Metadata
AA Openness IndexArtificial Analysis5.6†–

Data as of 2026-11-10. Best configuration of each model; every score links to its source on the model pages.

What changed

  • Gemini 4 Carbon (high): 2 new results (FrontierSWE, Vals RSI Index).
  • Gemini 4 Carbon (high): BenchLeader Index 69.3 → 68.5.
  • Gemini 4 Carbon (high): BenchLeader Index 68.8 → 69.3.
  • Gemini 3 Pro: 1 new result (Arena Search).
  • Gemini 4 Carbon (high): 1 score revision.
  • Gemini 4 Carbon (high): 2 score revisions.
  • Gemini 4 Carbon (high): 1 score revision.
  • Gemini 4 Carbon (high): 2 score revisions.
  • Gemini 4 Carbon (high): 1 score revision.
  • Gemini 3 Pro: 1 score revision.

Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs

Gemini 3 Pro vs Gemini 4 Carbon: questions

Is Gemini 4 Carbon better than Gemini 3 Pro?
Gemini 4 Carbon (high) leads on quality: 68.5 vs 60.2. The BenchLeader Index combines every independent quality benchmark; Gemini 4 Carbon (high) is ahead overall as of 10 Nov 2026, but check the category scores for your use.
Is Gemini 3 Pro or Gemini 4 Carbon better for coding?
Gemini 4 Carbon scores higher in coding (75 vs 58 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1441 for Gemini 3 Pro against 1768 for Gemini 4 Carbon.
Is Gemini 3 Pro or Gemini 4 Carbon better for agents and tool use?
Gemini 4 Carbon scores higher in agentic tasks (78 vs 61 on the category index, where 50 is average).
Do independent benchmarks agree on Gemini 3 Pro vs Gemini 4 Carbon?
Across the 7 benchmarks both have, Gemini 3 Pro leads 0, Gemini 4 Carbon leads 7. Every overall index that rates both puts Gemini 4 Carbon ahead.
When was Gemini 4 Carbon released?
Gemini 4 Carbon was released on 28 Oct 2026.