BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases4

Gemini 3.1 Pro vs o3

Full profiles: Gemini 3.1 Pro (Google) and o3 (OpenAI): every benchmark, price and rank.

Updated · Data from 18 sources

As of 2 Nov 2026, Gemini 3.1 Pro and o3 are neck and neck on quality (60.6 vs 61.0 on the BenchLeader Index, inside the ±5.3 margin), and o3 is 1.3× cheaper at $3.50 vs $4.50 per 1M tokens blended. Choose Gemini 3.1 Pro for multimodal (65 vs 57), reasoning (66 vs 61), and long context (66 vs 62); choose o3 for fast first answers (6.8 s vs 17.7 s) and agents & tools (70 vs 57), maths (71 vs 61), and coding (62 vs 57).

Key facts (as of 2 Nov 2026)

  • Gemini 3.1 Pro scores 60.6 on the BenchLeader Index and o3 scores 61.0 (BenchLeader, 2 Nov 2026).
  • Gemini 3.1 Pro costs $4.50 per 1M tokens blended (list price, 2 Nov 2026).
  • o3 costs $3.50 per 1M tokens blended (list price, 2 Nov 2026).
  • Of the 24 benchmarks both models have, Gemini 3.1 Pro leads 21 (BenchLeader, 2 Nov 2026).
Gemini 3.1 Pro benchmarks Google o3 benchmarks OpenAIAdd 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
Codingo362 vs 57 category score; Arena Coding 1460 vs 1521Medium
Agents & tool useo370 vs 57 category score; Terminal-Bench Hard 37.1% vs 53.8%High
Mathso371 vs 61 category score; Arena Maths 1448 vs 1488High
ReasoningGemini 3.1 Pro66 vs 61 category score; Arena Hard Prompts 1508 vs 1441Medium
KnowledgeEither56 vs 56 category score; PRBench Finance 41.9% vs 47.7%Either
Long documentsGemini 3.1 Pro66 vs 62 category score; AA-LCR v1.1 82.0% vs 74.7%Medium
Images and other mediaGemini 3.1 Pro65 vs 57 category score; Arena Vision 1296 vs 1214High
Following instructionsGemini 3.1 Pro67 vs 65 category score; IFBench 77.1% vs 71.4%Medium
Chat quality (human preference)o361 vs 58 on Arena votesMedium
Lowest costo3$3.50 vs $4.50 per 1M blended; agentic coding session $0.152 vs $0.168–
Fastest first answero36.8 s (o3) vs 17.7 s (Gemini 3.1 Pro) 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 o3 are neck and neck on quality (60.6 vs 61.0, within the ±5.3 margin).
  • Gemini 3.1 Pro is stronger in instruction following, long context, multimodal, reasoning.
  • o3 is stronger in agents & tools, coding, human preference, knowledge, maths.
  • o3 is 1.3× cheaper ($3.50 vs $4.50 per 1M blended).

At a glance

SpecGemini 3.1 Proo3
DeveloperGoogleOpenAI
Released19 Feb 202616 Apr 2025
Context window (tokens)1,048,576200,000
Compared atGemini 3.1 Proo3
Open weightsNoNo
Price in / out / cached input per 1M$2.00 / $12.00 / $0.200$2.00 / $8.00 / $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 Proo3Leader
Reasoning66.360.9Gemini 3.1 Pro
Coding56.761.8o3
Agents & tools57.170.3o3
Maths60.770.5o3
Knowledge55.956.1level
Instruction following66.965.1Gemini 3.1 Pro
Human preference58.461.1o3
Multimodal64.656.7Gemini 3.1 Pro
Long context65.962.1Gemini 3.1 Pro

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 & toolsMathsKnowledgeInstructionfollowingLong 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, Instruction following: 66.9Gemini 3.1 Pro, Long context: 65.9Gemini 3.1 Pro, Multimodal: 64.6Gemini 3.1 Pro, Human preference: 58.4o3, Reasoning: 60.9o3, Coding: 61.8o3, Agents & tools: 70.3o3, Maths: 70.5o3, Knowledge: 56.1o3, Instruction following: 65.1o3, Long context: 62.1o3, Multimodal: 56.7o3, Human preference: 61.1
ModelIndex

Where sources disagree

Across the 24 benchmarks both have, Gemini 3.1 Pro leads 21, o3 leads 3. Every overall index that rates both puts Gemini 3.1 Pro ahead.

Overall indexGemini 3.1 Proo3Leader
BenchLeader Index60.661.0neck and neck
Arena Text14871432Gemini 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 Proo3
Blended price per 1M tokens$4.50$3.50
Input / output per 1M$2.00 / $12.00$2.00 / $8.00
Cached input per 1M$0.200$0.500
Price per task: chat reply$0.0044$0.0032
Price per task: summarize a 30-page report$0.031$0.029
Price per task: code edit$0.030$0.024
Price per task: agentic coding session$0.168$0.152
Price per task: structured extraction$0.0064$0.0056
Agentic coding session, input mostly cached$0.087$0.085
Output speed143 tok/s160 tok/s
Time to first answer17.7 s6.8 s
Time to a complete answer21.2 s9.9 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. o3 scores 61.0 and takes 7 s to its first answer token; o3 (high) scores 53.8 in 7 s, o3 (medium) scores 53.3 in 7 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

177 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 Proo3
BenchLeader IndexBenchLeader60.661.0
Agents & tools scoreBenchLeader57.170.3
Coding scoreBenchLeader56.761.8
Human preference scoreBenchLeader58.461.1
Instruction following scoreBenchLeader66.965.1
Knowledge scoreBenchLeader55.956.1
Long context scoreBenchLeader65.962.1
Maths scoreBenchLeader60.770.5
Multimodal scoreBenchLeader64.656.7
Reasoning scoreBenchLeader66.360.9
Blended price $/MBenchLeader$4.50$3.50
Output speedBenchLeader143 tok/s160 tok/s
Time to first answerBenchLeader17.7 s6.8 s
Context windowBenchLeader1,048,576 tokens200,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub94.1%81.8%†
Humanity's Last ExamScale AI / CAIS46.4%20.3%†
SimpleBenchSimpleBench79.6%53.1%†
Arena Hard PromptsArena (LMArena)15081441
LiveBench 1.0 ReasoningLiveBench84.0%†–
GPQA Diamond (AA)Artificial Analysis94.1%82.7%
Humanity's Last Exam (AA)Artificial Analysis47.0%20.1%
HLE-Diamond (AA)Artificial Analysis31.4%–
GPQA Diamond (Vals)Vals AI95.5%†84.1%†
EnigmaEvalScale AI SEAL36.8%†–
Kagi LLM BenchmarkKagi LLM Benchmark–67.6%
GPQA Diamond (HELM)HELM Capabilities–75.3%
ARC-AGI-1ARC Prize98.0%60.8%†
ARC-AGI-2ARC Prize77.1%6.5%†
ARC-AGI-3ARC Prize0.4%–
CritPtArtificial Analysis17.7%1.1%
MysteryMechanismVals AI20.7%†–
SciPredictScale AI SEAL–17.9%
Chess PuzzlesEpoch AI Benchmarking Hub51.0%†36.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub34.0%†29.0%†
CL-benchEpoch AI Benchmarking Hub20.8%17.8%†
CL-bench LifeEpoch AI Benchmarking Hub16.9%–
LMCAEpoch AI Benchmarking Hub53.8%†39.7%†
DTBenchEpoch AI Benchmarking Hub95.5%†84.8%†
ForecastBenchForecastBench64.3%64.6%
Show all 177 benchmarks
MetricPublisherGemini 3.1 Proo3
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–62.3%†
SciCodeSciCode58.7%–
WeirdMLWeirdML72.1%52.4%†
GSO-BenchGSO-Bench22.6%8.8%†
Arena CodingArena (LMArena)15211460
Arena WebDevArena (LMArena)1447–
LiveBench 1.0 CodingLiveBench76.5%†–
LiveCodeBenchVals AI88.5%†83.9%†
IOIVals AI51.8%†–
SWE-bench (Vals)Vals AI78.8%†–
SWE-Bench ProScale AI SEAL46.1%†–
Aider PolyglotAider polyglot leaderboard–81.3%
SWE-bench Verified (bash only)SWE-bench–58.4%
SWE-bench Verified (any scaffold)SWE-bench–58.4%
Code MigrationVals AI17.3%†–
ProgramBenchVals AI0.0%†–
Vibe Code Bench 1-100Vals AI6.7%†–
Vibe Code Bench v1.1Vals AI32.0%†–
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%†–
CadEvalEpoch AI Benchmarking Hub–74.0%†
DeepSWE v1.1Epoch AI Benchmarking Hub11.7%–
ALE-BenchEpoch AI Benchmarking Hub1160.6933.5†
AlgoTuneAlgoTune2.0–
Terminal-Bench 4.0 (AA)Artificial Analysis4.0%–
Terminal-Bench 2.1 (AA)Artificial Analysis73.8%–
GameBench 2 OverallDuelLab GameBench 2–36.2
GameBench 2 by reasoning settingDuelLab GameBench 2–32.0†
LiveCodeBench (AA)Artificial Analysis–80.8%
Agents & tools
Terminal-BenchTerminal-Bench80.2%–
GDPvalOpenAI–30.8%†
APEX-AgentsMercor35.1%–
Arena AgentArena (LMArena)-4.5–
LiveBench Agentic CodingLiveBench44.1%†–
Terminal-Bench HardArtificial Analysis53.8%37.1%
τ²-Bench Telecom (AA)Artificial Analysis95.6%80.7%
Terminal-Bench 2.1 (Vals)Vals AI70.8%†–
MCP AtlasScale AI SEAL78.2%†–
HiL-BenchScale AI SEAL35.3%–
BFCL OverallBerkeley Function Calling Leaderboard–63.0%
τ³-bench Bankingτ²-bench26.0%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis14.6%–
GDPval-AA v2.1Artificial Analysis799–
τ³-Banking (AA)Artificial Analysis21.4%–
ITBench SRE (AA)Artificial Analysis30.3%–
Analyst Agent (AA)Artificial Analysis41.3%–
APEX-Agents (AA)Artificial Analysis32.0%–
Finance Agent v2Vals AI43.0%†–
Harvey's Legal Agent Benchmark v1Vals AI0.0%†–
Legal Research BenchVals AI20.7%†–
SREBenchVals AI0.0%†–
Tax Agent BenchVals AI9.3%†–
Terminal-Bench 2.0 (Vals)Vals AI67.4%†–
Terminal-Bench 4.0 (Vals)Vals AI2.5%†–
Terminal-Bench ScienceVals AI1.4%†–
BALROGBALROG57.0%–
DeepResearch BenchDeepResearch Bench47.8%†45.2%†
PostTrainBenchEpoch AI Benchmarking Hub22.0%–
ExploitBenchEpoch AI Benchmarking Hub26.1%–
METR Time HorizonsMETR77.0%63.6%
GBAEvalEpoch AI Benchmarking Hub0.8%–
Vending-Bench 2Andon Labs911.2–
AutomationBenchArtificial Analysis35.4%–
GDP.pdf (AA)Artificial Analysis17.8%–
Harvey LAB-AA v1.1Harvey0.0%–
EnterpriseOps-GymArtificial Analysis42.2%–
AA-Briefcase v1.1Artificial Analysis457–
BrowseComp-PlusBrowseComp-Plus–50.5%
BrowseCompOpenAI31.2%–
DeepSearchQAGoogle60.2%–
FACTS SearchGoogle83.7%–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub59.6%33.3%†
FrontierMath Tier 4Epoch AI Benchmarking Hub26.8%–
OTIS Mock AIMEEpoch AI Benchmarking Hub95.6%83.9%†
MATH Level 5Epoch AI Benchmarking Hub–97.8%†
ProofBenchVals AI26.0%–
LiveBench MathematicsLiveBench91.0%†–
AIME (Vals)Vals AI98.1%†85.3%†
MGSMVals AI–91.8%†
AIME 2026MathArena98.3%–
HMMT February 2026MathArena94.7%–
USAMO 2026MathArena74.4%–
IMO 2025MathArena–16.7%†
MathArena ApexMathArena60.9%–
Omni-MATH (HELM)HELM Capabilities–71.4%
MATH 500Vals AI–94.6%†
Arena MathsArena (LMArena)14881448
AIME 2025 (AA)Artificial Analysis–88.3%
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub73.5%†49.4%†
LiveBench 1.0 Data AnalysisLiveBench78.5%†–
MMLU-ProVals AI91.0%†85.6%†
LegalBenchVals AI87.4%†83.8%†
CorpFinVals AI64.5%†59.7%†
TaxEvalVals AI72.9%†74.6%†
MedQAVals AI96.4%†96.1%†
PRBench FinanceScale AI SEAL41.9%47.7%
PRBench LegalScale AI SEAL44.0%48.6%
MultiNRCScale AI SEAL64.7%45.5%†
MMLU-Pro (HELM)HELM Capabilities–85.9%
BioMysteryBenchVals AI60.0%†–
CaseLaw v2Vals AI64.8%†–
Excel Modeling BenchmarkVals AI52.6%†–
MedCodeVals AI59.1%†47.3%†
MedScribeVals AI76.1%†76.7%†
Public Benefits Bench v1.1Vals AI53.8%†–
DrugDiscoveryBenchScale AI SEAL41.9%†–
EBR-benchEpoch AI Benchmarking Hub14.3%–
GDP.pdfEpoch AI Benchmarking Hub17.0%–
Global-MMLU-Lite (AA)Artificial Analysis93.2%–
Instruction following
LiveBench LanguageLiveBench85.4%†–
LiveBench Instruction FollowingLiveBench79.1%†–
IFBenchArtificial Analysis77.1%71.4%
MultiChallengeScale AI SEAL71.4%56.6%†
TutorBenchScale AI SEAL53.0%52.1%†
IFEval (HELM)HELM Capabilities–86.9%
WildBench (HELM)HELM Capabilities–86.1%
Arena Instruction FollowingArena (LMArena)14801403
Human preference
Arena TextArena (LMArena)14871432
EQ-Bench 4EQ-Bench1132–
Arena Creative WritingArena (LMArena)14811383
Arena Multi-turnArena (LMArena)14961420
Arena Longer QueriesArena (LMArena)14991411
Lech Mazur Creative WritingLech Mazur–83.9%†
Multimodal
Arena VisionArena (LMArena)12961214
MMMU-ProArtificial Analysis82.4%70.1%
VISTAScale AI SEAL–50.1%†
MMMU (validation)MMMU–82.9%
MMMU-Pro (official)MMMU80.5%†76.4%
MMMU-Pro (Vals)Vals AI88.2%†80.4%†
MortgageTaxVals AI69.4%†65.7%†
SAGEVals AI83.2%†–
VTBScale AI SEAL29.0%13.7%
GeoBenchGeoBench–60.0%†
VPCTEpoch AI Benchmarking Hub–52.0%†
Blueprint-Bench 2Epoch AI Benchmarking Hub26.5%–
Long context
Fiction.LiveBench 120kFiction.live–100.0%†
AA-LCR v1.1Artificial Analysis82.0%74.7%
Arena DocumentArena (LMArena)1459–
MLCRArtificial Analysis11.1%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub155.2146.9
LiveBench 1.0LiveBench77.0%†–
AA Intelligence Index v4.3.2Artificial Analysis29.720.2
AA Progress Index v1Artificial Analysis296.1258.3
HELM Capabilities meanHELM Capabilities–81.1%
Vals Science IndexVals AI27.4%†–
Vals Multimodal IndexVals AI56.1%†–
Safety & honesty
FORTRESSScale AI SEAL29.8%16.0%
MASKScale AI SEAL42.4%84.5%†
PropensityBenchScale AI SEAL–10.5%

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)).
  • o3: 1 new result (AA Progress Index v1).
  • 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.
  • Gemini 3.1 Pro: 2 score revisions.
  • Gemini 3.1 Pro: 1 score revision.
  • o3: 1 score revision.
  • Gemini 3.1 Pro: BenchLeader Index 62.3 → 59.3.

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

Gemini 3.1 Pro vs o3: questions

Is Gemini 3.1 Pro better than o3?
Gemini 3.1 Pro and o3 are neck and neck on quality (60.6 vs 61.0, within the ±5.3 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 o3: which is cheaper, and will that change?
o3 is cheaper today: $3.50 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 o3 better for coding?
o3 scores higher in coding (62 vs 57 on the category index, where 50 is average). On Arena Coding, the top shared coding benchmark, it is 1521 for Gemini 3.1 Pro against 1460 for o3.
Which is faster, Gemini 3.1 Pro or o3?
o3 streams faster (160 vs 143 output tokens per second, Artificial Analysis 7-day median).
Is Gemini 3.1 Pro or o3 better for agents and tool use?
o3 scores higher in agentic tasks (70 vs 57 on the category index, where 50 is average). On Terminal-Bench Hard, the top shared agents & tools benchmark, it is 53.8% for Gemini 3.1 Pro against 37.1% for o3.
More questions (4)
Which is better for long documents?
Gemini 3.1 Pro scores higher on long-context benchmarks (66 vs 62). Gemini 3.1 Pro accepts more context: 1,048,576 against 200,000 tokens for o3.
Which has lower latency?
o3 answers first: 6.8 s against 17.7 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 o3?
Across the 24 benchmarks both have, Gemini 3.1 Pro leads 21, o3 leads 3. Every overall index that rates both puts Gemini 3.1 Pro ahead.
When was Gemini 3.1 Pro released?
Gemini 3.1 Pro was released on 19 Feb 2026.