BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases4

Gemini 3.1 Pro vs Grok 4.6

Full profiles: Gemini 3.1 Pro (Google) and Grok 4.6 (SpaceXAI): every benchmark, price and rank.

Updated · Data from 12 sources

As of 2 Nov 2026, Gemini 3.1 Pro and Grok 4.6 are neck and neck on quality (60.6 vs 60.5 on the BenchLeader Index, inside the ±5.4 margin), and Grok 4.6 is 1.5× cheaper at $3.00 vs $4.50 per 1M tokens blended. Choose Gemini 3.1 Pro for multimodal (65 vs 60) and reasoning (66 vs 64) and fast first answers (17.7 s vs 40.0 s), where it also streams ~1.5× faster; choose Grok 4.6 for human preference (64 vs 58), coding (61 vs 57), and knowledge (60 vs 56).

Key facts (as of 2 Nov 2026)

  • Gemini 3.1 Pro scores 60.6 on the BenchLeader Index and Grok 4.6 scores 60.5 (BenchLeader, 2 Nov 2026).
  • Gemini 3.1 Pro costs $4.50 per 1M tokens blended (list price, 2 Nov 2026).
  • Grok 4.6 costs $3.00 per 1M tokens blended (list price, 2 Nov 2026).
  • Gemini 3.1 Pro streams ~1.5× faster than Grok 4.6 (Artificial Analysis, 7-day median, 2 Nov 2026).
  • Of the 33 benchmarks both models have, Gemini 3.1 Pro leads 19 (BenchLeader, 2 Nov 2026).
  • On Arena WebDev, Gemini 3.1 Pro is rated 1447 and Grok 4.6 1616 (LMArena, as published by 2 Nov 2026).
Gemini 3.1 Pro benchmarks Google Grok 4.6 benchmarks high SpaceXAIAdd a third model

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
CodingGrok 4.661 vs 57 category score; Arena WebDev 1616 vs 1447Medium
Agents & tool useGrok 4.658 vs 57 category score; τ³-Banking (AA) 50.7% vs 21.4%Medium
MathsGrok 4.663 vs 61 category score; OTIS Mock AIME 97.8% vs 95.6%Medium
ReasoningGemini 3.1 Pro66 vs 64 category score; GPQA Diamond 94.1% vs 94.0%Medium
KnowledgeGrok 4.660 vs 56 category score; GDP.pdf 16.0% vs 17.0%Medium
Long documentsEither66 vs 65 category score; AA-LCR v1.1 82.0% vs 80.3%Either
Images and other mediaGemini 3.1 Pro65 vs 60 category score; Arena Vision 1296 vs 1263Medium
Chat quality (human preference)Grok 4.664 vs 58 on Arena votesMedium
Lowest costGrok 4.6$3.00 vs $4.50 per 1M blended; agentic coding session $0.144 vs $0.168–
Fastest first answerGemini 3.1 Pro17.7 s (Gemini 3.1 Pro) vs 40.0 s (Grok 4.6 (high)) 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 3.1 Pro and Grok 4.6 (high) are neck and neck on quality (60.6 vs 60.5, within the ±5.4 margin).
  • Gemini 3.1 Pro is stronger in instruction following, long context, multimodal, reasoning.
  • Grok 4.6 (high) is stronger in agents & tools, coding, human preference, knowledge, maths.
  • Grok 4.6 (high) is 1.5× cheaper ($3.00 vs $4.50 per 1M blended).
  • Gemini 3.1 Pro streams 2.1× faster (143 vs 70 tokens per second).

At a glance

SpecGemini 3.1 ProGrok 4.6
DeveloperGoogleSpaceXAI
Released19 Feb 202611 Aug 2026
Context window (tokens)1,048,576500,000
Compared atGemini 3.1 ProGrok 4.6 (high)
Open weightsNoNo
Price in / out / cached input per 1M$2.00 / $12.00 / $0.200$2.00 / $6.00 / $0.500

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 3.1 ProGrok 4.6 (high)Leader
Reasoning66.364.0Gemini 3.1 Pro
Coding56.761.3Grok 4.6
Agents & tools57.158.1Grok 4.6
Maths60.763.3Grok 4.6
Knowledge55.959.6Grok 4.6
Instruction following66.9–not comparable
Human preference58.464.0Grok 4.6
Multimodal64.659.8Gemini 3.1 Pro
Long context65.965.0level

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 & toolsMathsKnowledgeLong contextMultimodalHumanpreferenceGemini 3.1 Pro, Reasoning: 66.3Gemini 3.1 Pro, Coding: 56.7Gemini 3.1 Pro, Agents & tools: 57.1Gemini 3.1 Pro, Maths: 60.7Gemini 3.1 Pro, Knowledge: 55.9Gemini 3.1 Pro, Long context: 65.9Gemini 3.1 Pro, Multimodal: 64.6Gemini 3.1 Pro, Human preference: 58.4Grok 4.6, Reasoning: 64.0Grok 4.6, Coding: 61.3Grok 4.6, Agents & tools: 58.1Grok 4.6, Maths: 63.3Grok 4.6, Knowledge: 59.6Grok 4.6, Long context: 65.0Grok 4.6, Multimodal: 59.8Grok 4.6, Human preference: 64.0

Not drawn, measured for fewer than two of these models: Instruction following.

ModelIndex

Where sources disagree

Across the 33 benchmarks both have, Gemini 3.1 Pro leads 19, Grok 4.6 leads 13 and 1 is a tie. Every overall index that rates both puts Gemini 3.1 Pro ahead.

Overall indexGemini 3.1 ProGrok 4.6 (high)Leader
BenchLeader Index60.660.5neck and neck
Arena Text14871456Gemini 3.1 Pro

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 3.1 ProGrok 4.6 (high)
Blended price per 1M tokens$4.50$3.00
Input / output per 1M$2.00 / $12.00$2.00 / $6.00
Cached input per 1M$0.200$0.500
Price per task: chat reply$0.0044$0.0026
Price per task: summarize a 30-page report$0.031$0.028
Price per task: code edit$0.030$0.021
Price per task: agentic coding session$0.168$0.144
Price per task: structured extraction$0.0064$0.0052
Agentic coding session, input mostly cached$0.087$0.076
Output speed143 tok/s70 tok/s
Time to first answer17.7 s40.0 s
Time to a complete answer21.2 s47.2 s

Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.1 Pro scores 60.6 and takes 18 s to its first answer token; Gemini 3.1 Pro (high) scores 60.3 in 18 s. Grok 4.6 (high) scores 60.5 and takes 40 s to its first answer token; Grok 4.6 (medium) scores 59.6 in 35 s, Grok 4.6 (xhigh) scores 58.4 in 52 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

165 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 3.1 ProGrok 4.6 (high)
BenchLeader IndexBenchLeader60.660.5
Agents & tools scoreBenchLeader57.158.1
Coding scoreBenchLeader56.761.3
Human preference scoreBenchLeader58.464.0
Instruction following scoreBenchLeader66.9–
Knowledge scoreBenchLeader55.959.6
Long context scoreBenchLeader65.965.0
Maths scoreBenchLeader60.763.3
Multimodal scoreBenchLeader64.659.8
Reasoning scoreBenchLeader66.364.0
Blended price $/MBenchLeader$4.50$3.00
Output speedBenchLeader143 tok/s70 tok/s
Time to first answerBenchLeader17.7 s40.0 s
Context windowBenchLeader1,048,576 tokens500,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub94.1%94.0%
Humanity's Last ExamScale AI / CAIS46.4%–
SimpleBenchSimpleBench79.6%75.9%†
Arena Hard PromptsArena (LMArena)15081481
LiveBench 2.0 ReasoningLiveBench–91.2%†
LiveBench 1.0 ReasoningLiveBench84.0%†90.5%†
GPQA Diamond (AA)Artificial Analysis94.1%95.0%
Humanity's Last Exam (AA)Artificial Analysis47.0%42.9%
HLE-Diamond (AA)Artificial Analysis31.4%–
GPQA Diamond (Vals)Vals AI95.5%†94.7%
EnigmaEvalScale AI SEAL36.8%†–
ARC-AGI-1ARC Prize98.0%87.0%
ARC-AGI-2ARC Prize77.1%65.1%
ARC-AGI-3ARC Prize0.4%2.1%†
CritPtArtificial Analysis17.7%17.1%
MysteryMechanismVals AI20.7%†30.6%
Chess PuzzlesEpoch AI Benchmarking Hub51.0%†33.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub34.0%†34.0%†
CL-benchEpoch AI Benchmarking Hub20.8%–
CL-bench LifeEpoch AI Benchmarking Hub16.9%–
LMCAEpoch AI Benchmarking Hub53.8%†48.5%†
DTBenchEpoch AI Benchmarking Hub95.5%†97.3%†
ForecastBenchForecastBench64.3%65.5%†
Coding
SciCodeSciCode58.7%56.5%
WeirdMLWeirdML72.1%67.3%
Show all 165 benchmarks
MetricPublisherGemini 3.1 ProGrok 4.6 (high)
Coding
FrontierCodeCognition–48.0%†
GSO-BenchGSO-Bench22.6%–
Arena CodingArena (LMArena)15211509
Arena WebDevArena (LMArena)14471616
LiveBench 2.0 CodingLiveBench–89.1%†
LiveBench 1.0 CodingLiveBench76.5%†76.8%†
LiveCodeBenchVals AI88.5%†88.2%
IOIVals AI51.8%†47.6%
SWE-bench (Vals)Vals AI78.8%†95.6%
SWE-Bench ProScale AI SEAL46.1%†–
Code MigrationVals AI17.3%†44.6%
ProgramBenchVals AI0.0%†1.0%
Vibe Code Bench 1-100Vals AI6.7%†14.8%
Vibe Code Bench v1.1Vals AI32.0%†76.2%
SWE-Bench Pro (private)Scale AI SEAL46.1%†–
SWE Atlas: Codebase QnAScale AI SEAL13.5%–
SWE Atlas: RefactoringScale AI SEAL33.8%–
SWE Atlas: Test WritingScale AI SEAL29.8%–
MirrorCodeEpoch AI Benchmarking Hub8.9%†41.1%†
DeepSWE v1.1Epoch AI Benchmarking Hub11.7%65.2%
CursorBenchCursor–40.4%
ALE-BenchEpoch AI Benchmarking Hub1160.61508.1†
AlgoTuneAlgoTune2.0–
FrontierSWEEpoch AI Benchmarking Hub–25.3%†
Terminal-Bench 4.0 (AA)Artificial Analysis4.0%21.2%
Terminal-Bench 2.1 (AA)Artificial Analysis73.8%88.4%
Startrise FrontendStartrise–72.1%†
GameBench 2 OverallDuelLab GameBench 2–61.0†
GameBench 2 by reasoning settingDuelLab GameBench 2–71.4†
Agents & tools
Terminal-BenchTerminal-Bench80.2%20.3%
APEX-AgentsMercor35.1%65.3%†
Arena AgentArena (LMArena)-4.50.5†
LiveBench Long AgenticLiveBench–28.0%†
LiveBench BrowserLiveBench–49.4%†
LiveBench Agentic CodingLiveBench44.1%†57.0%†
Terminal-Bench HardArtificial Analysis53.8%–
τ²-Bench Telecom (AA)Artificial Analysis95.6%–
Terminal-Bench 2.1 (Vals)Vals AI70.8%†78.3%
MCP AtlasScale AI SEAL78.2%†–
HiL-BenchScale AI SEAL35.3%–
τ³-bench Bankingτ²-bench26.0%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis14.6%–
GDPval-AA v2.1Artificial Analysis799–
τ³-Banking (AA)Artificial Analysis21.4%50.7%
ITBench SRE (AA)Artificial Analysis30.3%–
Analyst Agent (AA)Artificial Analysis41.3%41.3%
APEX-Agents (AA)Artificial Analysis32.0%–
CyberBenchVals AI–67.7%
Finance Agent v2Vals AI43.0%†53.7%
Harvey's Legal Agent Benchmark v1Vals AI0.0%†15.8%
Legal Research BenchVals AI20.7%†48.1%
SkillsBenchVals AI–55.8%
SREBenchVals AI0.0%†–
Tax Agent BenchVals AI9.3%†33.1%
Terminal-Bench 2.0 (Vals)Vals AI67.4%†–
Terminal-Bench 4.0 (Vals)Vals AI2.5%†17.2%
Terminal-Bench ScienceVals AI1.4%†11.4%
Time Horizon Index: KSPVals AI–5.8%
BALROGBALROG57.0%–
DeepResearch BenchDeepResearch Bench47.8%†–
PostTrainBenchEpoch AI Benchmarking Hub22.0%–
ExploitBenchEpoch AI Benchmarking Hub26.1%–
METR Time HorizonsMETR77.0%–
GBAEvalEpoch AI Benchmarking Hub0.8%–
Vending-Bench 2Andon Labs911.29047.0†
AutomationBenchArtificial Analysis35.4%66.7%
GDP.pdf (AA)Artificial Analysis17.8%17.0%
Harvey LAB-AA v1.1Harvey0.0%–
EnterpriseOps-GymArtificial Analysis42.2%48.3%
AA-Briefcase v1.1Artificial Analysis4571529
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–6.2%
BrowseCompOpenAI31.2%–
DeepSearchQAGoogle60.2%–
FACTS SearchGoogle83.7%–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub59.6%66.0%†
FrontierMath Tier 4Epoch AI Benchmarking Hub26.8%31.7%†
OTIS Mock AIMEEpoch AI Benchmarking Hub95.6%97.8%
ProofBenchVals AI26.0%51.0%†
LiveBench MathematicsLiveBench91.0%†92.6%†
AIME (Vals)Vals AI98.1%†–
AIME 2026MathArena98.3%–
HMMT February 2026MathArena94.7%–
USAMO 2026MathArena74.4%–
MathArena ApexMathArena60.9%–
Arena MathsArena (LMArena)14881443
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub73.5%†49.3%
LiveBench 2.0 Data AnalysisLiveBench–75.3%†
LiveBench ResearchLiveBench–84.5%†
LiveBench 1.0 Data AnalysisLiveBench78.5%†73.9%†
MMLU-ProVals AI91.0%†89.4%
LegalBenchVals AI87.4%†86.3%
CorpFinVals AI64.5%†66.2%
TaxEvalVals AI72.9%†71.1%
MedQAVals AI96.4%†–
PRBench FinanceScale AI SEAL41.9%–
PRBench LegalScale AI SEAL44.0%–
MultiNRCScale AI SEAL64.7%–
BioMysteryBenchVals AI60.0%†72.2%
CaseLaw v2Vals AI64.8%†–
Excel Modeling BenchmarkVals AI52.6%†62.7%
MedCodeVals AI59.1%†44.7%
MedScribeVals AI76.1%†86.5%
Public Benefits Bench v1.1Vals AI53.8%†66.8%
DrugDiscoveryBenchScale AI SEAL41.9%†–
EBR-benchEpoch AI Benchmarking Hub14.3%30.5%†
GDP.pdfEpoch AI Benchmarking Hub17.0%16.0%
Global-MMLU-Lite (AA)Artificial Analysis93.2%–
Instruction following
LiveBench LanguageLiveBench85.4%†83.7%†
LiveBench Instruction FollowingLiveBench79.1%†71.9%†
IFBenchArtificial Analysis77.1%–
MultiChallengeScale AI SEAL71.4%–
TutorBenchScale AI SEAL53.0%–
Arena Instruction FollowingArena (LMArena)14801451
Human preference
Arena TextArena (LMArena)14871456
EQ-Bench 4EQ-Bench1132–
Arena Creative WritingArena (LMArena)14811442
Arena Multi-turnArena (LMArena)14961448
Arena Longer QueriesArena (LMArena)14991474
Multimodal
Arena VisionArena (LMArena)12961263
LiveBench DesignLiveBench–79.8%†
MMMU-ProArtificial Analysis82.4%–
MMMU-Pro (official)MMMU80.5%†–
MMMU-Pro (Vals)Vals AI88.2%†–
MortgageTaxVals AI69.4%†64.2%
SAGEVals AI83.2%†–
VTBScale AI SEAL29.0%–
Blueprint-Bench 2Epoch AI Benchmarking Hub26.5%33.2%†
Long context
AA-LCR v1.1Artificial Analysis82.0%80.3%
Arena DocumentArena (LMArena)14591461
MLCRArtificial Analysis11.1%15.6%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub155.2156.9†
LiveBench 2.0LiveBench–67.8%†
LiveBench 1.0LiveBench77.0%†78.0%†
AA Intelligence Index v4.3.2Artificial Analysis29.744.3
AA Progress Index v1Artificial Analysis296.1321.5
Vals RSI IndexVals AI–25.5%†
Vals Science IndexVals AI27.4%†38.1%
Vals Multimodal IndexVals AI56.1%†–
Safety & honesty
FORTRESSScale AI SEAL29.8%–
MASKScale AI SEAL42.4%–

Data as of 2026-11-02. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261026-cb69502a12d9 (https://benchmarks.duellab.org/releases/gb2-20261026-cb69502a12d9/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Gemini 3.1 Pro: BenchLeader Index 61.6 → 60.5.
  • Gemini 3.1 Pro: 2 new results (AA Progress Index v1, HLE-Diamond (AA)).
  • Grok 4.6 (high): 1 new result (AA Progress Index v1).
  • Grok 4.6 (high): BenchLeader Index 61.4 → 60.9.
  • Gemini 3.1 Pro: BenchLeader Index 59.3 → 61.6.
  • Gemini 3.1 Pro: Gemini 3.1 Pro cost per task moved from $1.30/M to $1.44/M (11% up).
  • Gemini 3.1 Pro: 2 score revisions.
  • Grok 4.6 (high): 2 new results (Harvey's Legal Agent Benchmark v1, Vals Science Index).
  • Gemini 3.1 Pro: 2 score revisions.
  • Grok 4.6 (high): 1 score revision.

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

Gemini 3.1 Pro vs Grok 4.6: questions

Is Gemini 3.1 Pro better than Grok 4.6?
Gemini 3.1 Pro and Grok 4.6 (high) are neck and neck on quality (60.6 vs 60.5, within the ±5.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2 Nov 2026, so check the category scores for your use.
Gemini 3.1 Pro vs Grok 4.6: which is cheaper, and will that change?
Grok 4.6 is cheaper today: $3.00 against $4.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Gemini 3.1 Pro or Grok 4.6 better for coding?
Grok 4.6 scores higher in coding (61 vs 57 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1447 for Gemini 3.1 Pro against 1616 for Grok 4.6.
Which is faster, Gemini 3.1 Pro or Grok 4.6?
Gemini 3.1 Pro streams ~1.5× faster (143 vs 70 output tokens per second, Artificial Analysis 7-day median).
Is Gemini 3.1 Pro or Grok 4.6 better for agents and tool use?
Grok 4.6 scores higher in agentic tasks (58 vs 57 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 80.2% for Gemini 3.1 Pro against 20.3% for Grok 4.6.
More questions (4)
Which is better for long documents?
They are level on long-context benchmarks (66 vs 65). Gemini 3.1 Pro accepts more context: 1,048,576 against 500,000 tokens for Grok 4.6.
Which has lower latency?
Gemini 3.1 Pro answers first: 17.7 s against 40.0 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 3.1 Pro vs Grok 4.6?
Across the 33 benchmarks both have, Gemini 3.1 Pro leads 19, Grok 4.6 leads 13 and 1 is a tie. Every overall index that rates both puts Gemini 3.1 Pro ahead.
When was Grok 4.6 released?
Grok 4.6 was released on 11 Aug 2026.