BenchLeader

Gemini 4 Carbon vs GPT-6.2 Sol

Full profiles: Gemini 4 Carbon (Google) and GPT-6.2 Sol (OpenAI): every benchmark, price and rank.

Updated · Data from 12 sources

As of 19 Nov 2026, Gemini 4 Carbon and GPT-6.2 Sol are neck and neck on quality (68.0 vs 64.6 on the BenchLeader Index, inside the ±3.4 margin), and they cost about the same at $4.00 vs $4.00 per 1M tokens blended. Choose Gemini 4 Carbon for knowledge (75 vs 65), agents & tools (69 vs 65), and reasoning (72 vs 70); choose GPT-6.2 Sol for coding (75 vs 73), where it also streams faster.

Key facts (as of 19 Nov 2026)

  • Gemini 4 Carbon scores 68.0 on the BenchLeader Index and GPT-6.2 Sol scores 64.6 (BenchLeader, 19 Nov 2026).
  • Gemini 4 Carbon costs $4.00 per 1M tokens blended (list price, 19 Nov 2026).
  • GPT-6.2 Sol costs $4.00 per 1M tokens blended (list price, 19 Nov 2026).
  • GPT-6.2 Sol streams faster than Gemini 4 Carbon (Artificial Analysis, 7-day median, 19 Nov 2026).
  • Of the 39 benchmarks both models have, Gemini 4 Carbon leads 28 (BenchLeader, 19 Nov 2026).
  • On Arena WebDev, Gemini 4 Carbon is rated 1769 and GPT-6.2 Sol 1757 (LMArena, as published by 19 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
CodingGPT-6.2 Sol75 vs 73 category score; FrontierCode 2.0 (AA) 44.4% vs 40.6%Medium
Agents & tool useGemini 4 Carbon69 vs 65 category score; Terminal-Bench 64.5% vs 61.5%Medium
ReasoningGemini 4 Carbon72 vs 70 category score; HLE-Diamond (AA) 69.1% vs 64.0%Medium
KnowledgeGemini 4 Carbon75 vs 65 category score; AA-Omniscience v1.1 53.5 vs 40.4High
Chat quality (human preference)–Only Gemini 4 Carbon has an Arena rating (73 category score)Not comparable
Lowest costGPT-6.2 Sol$4.00 vs $8.00 per 1M blended; agentic coding session $0.160 vs $0.320–
Fastest first answerEither72.4 s (Gemini 4 Carbon (high)) vs 78.8 s (GPT-6.2 Sol (max)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Gemini 4 Carbon (high) and GPT-6.2 Sol (max) are neck and neck on quality (68.0 vs 64.6, within the ±3.4 margin).
  • Gemini 4 Carbon (high) is stronger in agents & tools, composite, knowledge, reasoning.
  • GPT-6.2 Sol (max) is stronger in coding.
  • They cost about the same ($4.00 per 1M blended).
  • GPT-6.2 Sol (max) streams 1.4× faster (74 vs 55 tokens per second).

At a glance

SpecGemini 4 CarbonGPT-6.2 Sol
DeveloperGoogleOpenAI
Released28 Oct 202617 Nov 2026
Context window (tokens)1,048,5761,050,000
Compared atGemini 4 Carbon (high)GPT-6.2 Sol (max)
Open weightsNoNo
Price in / out / cached input per 1M$4.00 / $20.00 / $0.100$2.00 / $10.00 / $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 4 Carbon (high)GPT-6.2 Sol (max)Leader
Reasoning71.870.1Gemini 4 Carbon
Coding72.874.5GPT-6.2 Sol
Agents & tools69.165.1Gemini 4 Carbon
Maths71.3–not comparable
Knowledge74.765.4Gemini 4 Carbon
Human preference72.6–not comparable
Long context64.3–not comparable
Composite67.864.5Gemini 4 Carbon

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 & toolsKnowledgeGemini 4 Carbon, Reasoning: 71.8Gemini 4 Carbon, Coding: 72.8Gemini 4 Carbon, Agents & tools: 69.1Gemini 4 Carbon, Knowledge: 74.7GPT-6.2 Sol, Reasoning: 70.1GPT-6.2 Sol, Coding: 74.5GPT-6.2 Sol, Agents & tools: 65.1GPT-6.2 Sol, Knowledge: 65.4

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

Model, click to add or removeIndex

Where sources disagree

Across the 39 benchmarks both have, Gemini 4 Carbon leads 28, GPT-6.2 Sol leads 9 and 2 are tied. Every overall index that rates both puts Gemini 4 Carbon ahead.

Overall indexGemini 4 Carbon (high)GPT-6.2 Sol (max)Leader
BenchLeader Index68.064.6neck and neck
AA Intelligence Index v5.0.160.057.3Gemini 4 Carbon
Vals Index70.566.1Gemini 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 4 Carbon (high)GPT-6.2 Sol (max)
Blended price per 1M tokens$8.00$4.00
Input / output per 1M$4.00 / $20.00$2.00 / $10.00
Cached input per 1M$0.100$0.100
Price per task: chat reply$0.0076$0.0038
Price per task: summarize a 30-page report$0.060$0.030
Price per task: code edit$0.054$0.027
Price per task: agentic coding session$0.320$0.160
Price per task: structured extraction$0.012$0.0060
Agentic coding session, input mostly cached$0.144$0.074
Cost per benchmark task (AA Index)$7.76$1.01
Output tokens per task (AA Index)123k40k
Of which reasoning72k23k
Time per task (AA Index)38 min9 min
Output speed55 tok/s74 tok/s
Time to first answer72.4 s78.8 s
Time to a complete answer, last measured 13 Nov 2026160.6 s–

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $7.76 on Gemini 4 Carbon (high) and $1.01 on GPT-6.2 Sol (max), and Gemini 4 Carbon (high) wrote 3.0× as many output tokens for the same work, which a per-token price hides.

Each model is compared at its best-ranked reasoning-effort setting. Gemini 4 Carbon (high) scores 68.0 and takes 72 s to its first answer token; Gemini 4 Carbon (medium) scores 64.7 in 33 s, Gemini 4 Carbon (low) scores 59.0 in 9 s. GPT-6.2 Sol (max) scores 64.6 and takes 79 s to its first answer token; GPT-6.2 Sol (xhigh) scores 63.3 in 32 s, GPT-6.2 Sol (high) scores 62.3 in 12 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

109 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 4 Carbon (high)GPT-6.2 Sol (max)
BenchLeader IndexBenchLeader68.064.6
Agents & tools scoreBenchLeader69.165.1
Coding scoreBenchLeader72.874.5
Composite scoreBenchLeader67.864.5
Human preference scoreBenchLeader72.6–
Knowledge scoreBenchLeader74.765.4
Long context scoreBenchLeader64.3–
Maths scoreBenchLeader71.3–
Reasoning scoreBenchLeader71.870.1
Blended price $/MBenchLeader$8.00$4.00
Output speedBenchLeader55 tok/s74 tok/s
Time to first answerBenchLeader72.4 s78.8 s
Context windowBenchLeader1,048,576 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub96.5%–
SimpleBenchSimpleBench92.6%†–
Arena Hard PromptsArena (LMArena)1546–
LiveBench 2.0 ReasoningLiveBench90.6%90.6%
HLE-Diamond (AA)Artificial Analysis69.1%64.0%
Kagi LLM BenchmarkKagi LLM Benchmark–67.0%†
ARC-AGI-1ARC Prize97.5%97.5%
ARC-AGI-2ARC Prize92.1%85.8%
ARC-AGI-3ARC Prize96.9%98.2%
CritPt v2 hard subset (AA)Artificial Analysis69.0%62.5%
MysteryMechanismVals AI–36.9%
HLE-Diamond (Scale)Scale AI SEAL–56.5%†
Arena ExpertArena (LMArena)1566–
Chess PuzzlesEpoch AI Benchmarking Hub78.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub73.0%–
Coding
SciCodeSciCode64.2%–
Arena CodingArena (LMArena)1567–
Arena WebDevArena (LMArena)17691757
LiveBench 2.0 CodingLiveBench91.0%83.7%
Vibe Code Bench v1.1Vals AI89.1%87.6%
Vibe Code Bench 1-100 v1.1Vals AI49.7%40.9%
SWE Atlas v1.1Scale AI SEAL71.5%†60.5%†
SWE Atlas v1.1: Codebase QnAScale AI SEAL73.4%†61.3%†
SWE Atlas v1.1: Test WritingScale AI SEAL76.7%†64.4%†
SWE Atlas v1.1: RefactoringScale AI SEAL64.3%†55.7%†
Show all 109 benchmarks
MetricPublisherGemini 4 Carbon (high)GPT-6.2 Sol (max)
Coding
CursorBenchCursor60.4%–
ALE-BenchEpoch AI Benchmarking Hub2374.7–
FrontierSWEEpoch AI Benchmarking Hub61.2%–
Terminal-Bench 4.0 (AA)Artificial Analysis65.2%60.1%
Design Arena WebsiteDesign Arena1304†–
FrontierCode 2.0 (AA)Artificial Analysis40.6%44.4%
Agents & tools
Terminal-BenchTerminal-Bench64.5%61.5%
APEX-AgentsMercor84.2%†–
Arena AgentArena (LMArena)16.111.3
LiveBench Long AgenticLiveBench71.8%63.6%
LiveBench BrowserLiveBench80.0%88.2%
HiL-BenchScale AI SEAL50.3%†–
GDPval-AA v2.1Artificial Analysis1733–
Analyst Agent (AA)Artificial Analysis–53.8%
Finance Agent v2.1Vals AI67.2%58.9%
Harvey's Legal Agent Benchmark v1.1Vals AI17.5%12.1%
Legal Research BenchVals AI55.8%37.0%
Tax Agent BenchVals AI51.8%24.8%
Terminal-Bench 4.0 (Vals)Vals AI63.1%60.6%
Terminal-Bench ScienceVals AI–58.6%
Arena Agent Task CompletionArena (LMArena)17.210.9
PostTrainBenchEpoch AI Benchmarking Hub41.0%–
Vending-Bench 2Andon Labs16484.2†–
AutomationBenchArtificial Analysis77.8%71.3%
GDP.pdf (AA)Artificial Analysis28.4%30.0%
Harvey LAB-AA v1.1Harvey–11.1%
EnterpriseOps-GymArtificial Analysis–28.9%
AA-Briefcase v1.1Artificial Analysis16111612
GDPxGDPx by SemiAnalysis10.9%–
Design Arena AgenticDesign Arena1304†–
Terminal-Bench-Science 0.1 (AA)Artificial Analysis64.8%67.6%
CWE-Bench (AA)Artificial Analysis–61.7%
DeepSec-Bench (AA)Artificial Analysis–51.6%
CyberGym end-to-end (AA)Artificial Analysis–0.0%
MKA (AA)Artificial Analysis54.3%–
FACTS SearchGoogle94.4%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub90.5%–
FrontierMath Tier 4Epoch AI Benchmarking Hub95.1%–
OTIS Mock AIMEEpoch AI Benchmarking Hub100.0%–
ArXivMathMathArena94.3%–
BrokenArXivMathArena84.4%–
Autoform-HardVals AI71.4%85.7%
FrontierMath ErdősEpoch AI Benchmarking Hub2.9%–
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub77.7%–
LiveBench 2.0 Data AnalysisLiveBench86.5%81.1%
LiveBench ResearchLiveBench92.5%90.1%
AA-Omniscience v1.1Artificial Analysis53.540.4
PRBench LegalScale AI SEAL53.4%†43.3%†
BioMysteryBenchVals AI–77.8%
Excel Modeling Benchmark v1.1Vals AI76.5%74.3%
DrugDiscoveryBenchScale AI SEAL71.5%62.2%†
EBR-benchEpoch AI Benchmarking Hub87.6%–
GDP.pdfEpoch AI Benchmarking Hub56.4%–
AA-Omniscience v1.1: accuracyArtificial Analysis58.3%61.6%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis88.6%44.6%
FACTS ParametricGoogle81.6%†–
Instruction following
MultiChallengeScale AI SEAL–72.0%†
Arena Instruction FollowingArena (LMArena)1524–
Human preference
Arena TextArena (LMArena)1526–
Arena Creative WritingArena (LMArena)1515–
Arena Multi-turnArena (LMArena)1563–
Arena Longer QueriesArena (LMArena)1537–
Multimodal
LiveBench DesignLiveBench80.2%80.0%
MMMU-ProArtificial Analysis89.3%†–
Humanity's Sixth SenseScale AI SEAL71.1%56.8%†
Blueprint-Bench 2Epoch AI Benchmarking Hub58.6%†–
Furniture AssemblyEpoch AI Benchmarking Hub77.5%–
VIALS-AAHandshake AI64.0%62.7%
AA-3DArtificial Analysis–1259
FACTS MultimodalGoogle60.2%†–
Long context
AA-LCR v1.1Artificial Analysis79.3%–
MLCRArtificial Analysis19.4%37.2%
FACTS GroundingGoogle77.2%†–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub169.8†–
LiveBench 2.0LiveBench83.3%80.3%
AA Intelligence Index v5.0.1Artificial Analysis60.057.3
AA Progress Index v1Artificial Analysis350.8–
Vals IndexVals AI70.566.1
Vals Finance IndexVals AI67.0%56.7%
Vals RSI IndexVals AI27.7%–
Vals Science IndexVals AI–57.8%
Vals Legal IndexVals AI75.9%66.7%
AA Cyber IndexArtificial Analysis–37.8
Safety & honesty
FORTRESSScale AI SEAL16.4%†–

Data as of 2026-11-19. Best configuration of each model; every score links to its source on the model pages.

What changed

  • GPT-6.2 Sol (max): BenchLeader Index 63.2 → 64.6.
  • Gemini 4 Carbon (high): 2 score revisions; input price moved from $2.00/M to $4.00/M (100% up); output price moved from $10.00/M to $20.00/M (100% up); cost per task moved from $4.90/M to $7.76/M (58% up).
  • GPT-6.2 Sol (max): 4 new results (VIALS-AA, Arena WebDev, Arena Agent…).
  • GPT-6.2 Sol (max): BenchLeader Index 62.4 → 63.2.
  • Gemini 4 Carbon (high): cost per task moved from $5.48/M to $4.90/M (11% down).
  • GPT-6.2 Sol (max): 2 new results (Vibe Code Bench 1-100 v1.1, Terminal-Bench).
  • Gemini 4 Carbon (high): BenchLeader Index 68.9 → 68.3.
  • Gemini 4 Carbon (high): 3 new results (PostTrainBench, ALE-Bench, GDP.pdf).
  • Gemini 4 Carbon (high): 2 score revisions.
  • Gemini 4 Carbon (high): 8 new results (LiveBench 2.0, LiveBench Long Agentic, LiveBench 2.0 Reasoning…).

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

Gemini 4 Carbon vs GPT-6.2 Sol: questions

Is Gemini 4 Carbon better than GPT-6.2 Sol?
Gemini 4 Carbon (high) and GPT-6.2 Sol (max) are neck and neck on quality (68.0 vs 64.6, within the ±3.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 19 Nov 2026, so check the category scores for your use.
Gemini 4 Carbon vs GPT-6.2 Sol: which is cheaper, and will that change?
They cost the same: $4.00 per million tokens blended. No price change has been announced for either.
Is Gemini 4 Carbon or GPT-6.2 Sol better for coding?
GPT-6.2 Sol scores higher in coding (75 vs 73 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1769 for Gemini 4 Carbon against 1757 for GPT-6.2 Sol.
Which is faster, Gemini 4 Carbon or GPT-6.2 Sol?
GPT-6.2 Sol streams faster (74 vs 55 output tokens per second, Artificial Analysis 7-day median).
Is Gemini 4 Carbon or GPT-6.2 Sol better for agents and tool use?
Gemini 4 Carbon scores higher in agentic tasks (69 vs 65 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 64.5% for Gemini 4 Carbon against 61.5% for GPT-6.2 Sol.
More questions (4)
Which is better for long documents?
Both take about 1.1M tokens: Gemini 4 Carbon 1,048,576, GPT-6.2 Sol 1,050,000.
Which has lower latency?
Gemini 4 Carbon answers first: 72.4 s against 78.8 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on Gemini 4 Carbon vs GPT-6.2 Sol?
Across the 39 benchmarks both have, Gemini 4 Carbon leads 28, GPT-6.2 Sol leads 9 and 2 are tied. Every overall index that rates both puts Gemini 4 Carbon ahead.
When was GPT-6.2 Sol released?
GPT-6.2 Sol was released on 17 Nov 2026.