BenchLeader

Gemini 3 Pro vs GPT-5

Full profiles: Gemini 3 Pro (Google) and GPT-5 (OpenAI): every benchmark, price and rank.

Updated · Data from 17 sources

As of 16 Nov 2026, Gemini 3 Pro and GPT-5 are neck and neck on quality (60.0 vs 59.0 on the BenchLeader Index, inside the ±6.4 margin), and GPT-5 is 1.3× cheaper at $3.44 vs $4.50 per 1M tokens blended. Choose Gemini 3 Pro for reasoning (63 vs 48), multimodal (66 vs 57), and agents & tools (62 vs 58); choose GPT-5 for maths (65 vs 54) and instruction following (69 vs 61).

Key facts (as of 16 Nov 2026)

  • Gemini 3 Pro scores 60.0 on the BenchLeader Index and GPT-5 scores 59.0 (BenchLeader, 16 Nov 2026).
  • Gemini 3 Pro costs $4.50 per 1M tokens blended (list price, 16 Nov 2026).
  • GPT-5 costs $3.44 per 1M tokens blended (list price, 16 Nov 2026).
  • Of the 12 benchmarks both models have, Gemini 3 Pro leads 12 (BenchLeader, 16 Nov 2026).
  • On Arena WebDev, Gemini 3 Pro is rated 1441 and GPT-5 1417 (LMArena, as published by 16 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
CodingEither58 vs 57 category score; Arena WebDev 1441 vs 1417Either
Agents & tool useGemini 3 Pro62 vs 58 category score; Terminal-Bench 69.4% vs 49.6%Medium
MathsGPT-565 vs 54 category score; OTIS Mock AIME 87.2% vs 91.4%High
ReasoningGemini 3 Pro63 vs 48 category score; GPQA Diamond 92.6% vs 85.3%High
Images and other mediaGemini 3 Pro66 vs 57 category score; GeoBench 84.0% vs 81.0%High
Following instructionsGPT-569 vs 61 category scoreHigh
Chat quality (human preference)–Only Gemini 3 Pro has an Arena rating (68 category score)Not comparable
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Gemini 3 Pro and GPT-5 (medium) are neck and neck on quality (60.0 vs 59.0, within the ±6.4 margin).
  • Gemini 3 Pro is stronger in agents & tools, coding, multimodal, reasoning.
  • GPT-5 (medium) is stronger in instruction following, maths.
  • GPT-5 (medium) is 1.3× cheaper ($3.44 vs $4.50 per 1M blended).

At a glance

SpecGemini 3 ProGPT-5
DeveloperGoogleOpenAI
Released18 Nov 20257 Aug 2025
Context window (tokens)–400,000
Compared atGemini 3 ProGPT-5 (medium)
Open weightsNoNo
Price in / out / cached input per 1M–$1.25 / $10.00 / $0.125

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 ProGPT-5 (medium)Leader
Reasoning62.647.9Gemini 3 Pro
Coding57.857.4level
Agents & tools61.758.1Gemini 3 Pro
Maths54.065.4GPT-5
Knowledge55.9–not comparable
Instruction following61.169.4GPT-5
Human preference67.7–not comparable
Multimodal65.656.6Gemini 3 Pro
Long context–70.4not 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 & toolsMathsInstructionfollowingMultimodalGemini 3 Pro, Reasoning: 62.6Gemini 3 Pro, Coding: 57.8Gemini 3 Pro, Agents & tools: 61.7Gemini 3 Pro, Maths: 54.0Gemini 3 Pro, Instruction following: 61.1Gemini 3 Pro, Multimodal: 65.6GPT-5, Reasoning: 47.9GPT-5, Coding: 57.4GPT-5, Agents & tools: 58.1GPT-5, Maths: 65.4GPT-5, Instruction following: 69.4GPT-5, Multimodal: 56.6

Not drawn, measured for fewer than two of these models: Knowledge, Long context, Human preference.

Model, click to add or removeIndex

Where sources disagree

Across the 12 benchmarks both have, Gemini 3 Pro leads 12, GPT-5 leads 0. Every overall index that rates both puts Gemini 3 Pro ahead.

Overall indexGemini 3 ProGPT-5 (medium)Leader
BenchLeader Index60.059.0neck and neck
Arena Text14861435†Gemini 3 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 ProGPT-5 (medium)
Blended price per 1M tokens–$3.44
Input / output per 1M– / –$1.25 / $10.00
Cached input per 1M–$0.125
Price per task: chat reply–$0.0035
Price per task: summarize a 30-page report–$0.021
Price per task: code edit–$0.022
Price per task: agentic coding session–$0.115
Price per task: structured extraction–$0.0045
Agentic coding session, input mostly cached–$0.064
Output speed–82 tok/s
Time to first answer–22.9 s
Time to a complete answer, last measured 13 Nov 2026–66.2 s

Each model is compared at its best-ranked reasoning-effort setting. Gemini 3 Pro scores 60.0; Gemini 3 Pro (high) scores 60.0. GPT-5 (medium) scores 59.0 and takes 23 s to its first answer token; GPT-5 (high) scores 58.2 in 49 s, GPT-5 scores 56.6 in 5 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

128 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 ProGPT-5 (medium)
BenchLeader IndexBenchLeader60.059.0
Agents & tools scoreBenchLeader61.758.1
Coding scoreBenchLeader57.857.4
Human preference scoreBenchLeader67.7–
Instruction following scoreBenchLeader61.169.4
Knowledge scoreBenchLeader55.9–
Maths scoreBenchLeader54.065.4
Multimodal scoreBenchLeader65.656.6
Reasoning scoreBenchLeader62.647.9
Long context scoreBenchLeader–70.4
Blended price $/MBenchLeader–$3.44
Output speedBenchLeader–82 tok/s
Time to first answerBenchLeader–22.9 s
Context windowBenchLeader–400,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub92.6%85.3%
Humanity's Last ExamScale AI / CAIS37.5%25.3%†
SimpleBenchSimpleBench76.4%56.7%†
Arena Hard PromptsArena (LMArena)15021449†
GPQA Diamond (AA)Artificial Analysis90.8%†84.2%
Humanity's Last Exam (AA)Artificial Analysis39.7%†25.4%
GPQA Diamond (Vals)Vals AI91.7%†85.6%†
Kagi LLM BenchmarkKagi LLM Benchmark80.1%72.7%†
GPQA Diamond (HELM)HELM Capabilities80.3%79.1%†
ARC-AGI-1ARC Prize75.0%56.2%
ARC-AGI-2ARC Prize31.1%7.5%
CritPtArtificial Analysis9.1%†0.0%
CritPt v2 hard subset (AA)Artificial Analysis–11.5%†
SciPredictScale AI SEAL25.3%–
Arena ExpertArena (LMArena)14991445†
Chess PuzzlesEpoch AI Benchmarking Hub–32.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub–15.0%
CL-benchEpoch AI Benchmarking Hub15.8%–
LMCAEpoch AI Benchmarking Hub–40.0%†
DTBenchEpoch AI Benchmarking Hub–90.7%†
ForecastBenchForecastBench65.0%65.0%†
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub72.9%71.5%
SciCodeSciCode–42.9%†
WeirdMLWeirdML69.9%39.8%†
GSO-BenchGSO-Bench18.6%6.9%†
Show all 128 benchmarks
MetricPublisherGemini 3 ProGPT-5 (medium)
Coding
Arena CodingArena (LMArena)15171462†
Arena WebDevArena (LMArena)14411417
LiveCodeBenchVals AI86.4%85.9%†
SWE-bench (Vals)Vals AI76.4%†69.0%†
SWE-Bench ProScale AI SEAL43.3%41.8%†
Aider PolyglotAider polyglot leaderboard–86.7%
SWE-bench Verified (bash only)SWE-bench74.2%65.0%
SWE-bench Multilingual (bash only)SWE-bench68.7%–
SWE-bench Verified (any scaffold)SWE-bench77.4%74.4%†
Vibe Code Bench v1.1Vals AI14.3%†20.1%†
SWE-Bench Pro (private)Scale AI SEAL17.9%14.9%†
MirrorCodeEpoch AI Benchmarking Hub–4.4%†
ALE-BenchEpoch AI Benchmarking Hub1176.81162.5†
AlgoTuneAlgoTune1.81.7†
Terminal-Bench 2.1 (AA)Artificial Analysis–35.2%†
GameBench 2 by reasoning settingDuelLab GameBench 2–38.0
Design Arena WebsiteDesign Arena12441174†
LiveCodeBench (AA)Artificial Analysis91.8%†70.3%
Agents & tools
Terminal-BenchTerminal-Bench69.4%49.6%
GDPvalOpenAI40.3%34.8%
Remote Labor IndexScale AI / CAIS1.3%1.7%†
Terminal-Bench HardArtificial Analysis41.7%†37.9%
τ²-Bench Telecom (AA)Artificial Analysis87.1%†86.5%
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–925†
τ³-Banking (AA)Artificial Analysis–22.1%†
Poker AgentVals AI1078.9%1103.2%†
Terminal-Bench 2.0 (Vals)Vals AI55.1%†37.1%†
Arena SearchArena (LMArena)12011189†
BALROGBALROG58.1%32.8%†
DeepResearch BenchDeepResearch Bench46.3%†48.6%
METR Time HorizonsMETR71.0%69.6%
Vending-Bench 2Andon Labs5478.2–
Design Arena AgenticDesign Arena1080–
BrowseComp-PlusBrowseComp-Plus–57.6%†
BrowseCompOpenAI–20.1%†
DeepSearchQAGoogle–59.4%†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub–55.4%†
FrontierMath Tier 4Epoch AI Benchmarking Hub–21.9%†
OTIS Mock AIMEEpoch AI Benchmarking Hub91.4%87.2%
MATH Level 5Epoch AI Benchmarking Hub–97.9%
ProofBenchVals AI20.0%18.0%†
AIME (Vals)Vals AI96.7%†93.4%†
MGSMVals AI93.9%†92.8%†
AIME 2026MathArena91.7%–
HMMT February 2026MathArena86.4%–
IMO 2025MathArena–38.1%†
MathArena ApexMathArena23.4%1.0%†
Omni-MATH (HELM)HELM Capabilities55.6%64.7%†
MATH 500Vals AI96.4%†96.0%†
Arena MathsArena (LMArena)14751411†
AIME 2025 (AA)Artificial Analysis95.7%†91.7%
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub–50.1%†
MMLU-ProVals AI90.1%86.5%†
LegalBenchVals AI87.0%†86.0%†
CorpFinVals AI63.7%†61.1%†
TaxEvalVals AI72.6%†73.4%†
MedQAVals AI96.0%†96.3%†
PRBench FinanceScale AI SEAL39.2%51.3%†
PRBench LegalScale AI SEAL40.6%49.0%†
MultiNRCScale AI SEAL59.0%52.1%†
MMLU-Pro (HELM)HELM Capabilities90.3%86.3%†
CaseLaw v2Vals AI53.1%†66.5%†
MedCodeVals AI52.2%†49.6%†
MedScribeVals AI72.0%†83.7%†
EBR-benchEpoch AI Benchmarking Hub–12.7%†
Global-MMLU-Lite (AA)Artificial Analysis92.2%†90.4%
Instruction following
IFBenchArtificial Analysis70.4%†70.6%
MultiChallengeScale AI SEAL65.7%63.2%†
TutorBenchScale AI SEAL53.7%55.3%†
IFEval (HELM)HELM Capabilities87.6%87.5%†
WildBench (HELM)HELM Capabilities85.9%85.7%†
Arena Instruction FollowingArena (LMArena)14731415†
Human preference
Arena TextArena (LMArena)14861427†
EQ-Bench Creative Writing v3EQ-Bench15241626†
Arena Creative WritingArena (LMArena)14841389†
Arena Multi-turnArena (LMArena)14951449†
Arena Longer QueriesArena (LMArena)14911435†
Lech Mazur Creative WritingLech Mazur–86.0%
Multimodal
Arena VisionArena (LMArena)13051232†
MMMU-ProArtificial Analysis80.2%†74.3%
VISTAScale AI SEAL51.5%49.7%†
MMMU-Pro (Vals)Vals AI87.5%†81.5%†
MortgageTaxVals AI69.1%†65.5%†
AudioMCScale AI SEAL54.6%†–
AudioMC (text output)Scale AI SEAL54.6%†–
VTBScale AI SEAL26.9%17.0%†
GeoBenchGeoBench84.0%81.0%
VPCTEpoch AI Benchmarking Hub91.0%63.2%
Long context
Fiction.LiveBench 120kFiction.live–96.9%
AA-LCR v1.1Artificial Analysis76.0%†76.0%
Arena DocumentArena (LMArena)1451–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub153.0150†
AA Intelligence Index v4.3.2Artificial Analysis28.0†22.9
AA Progress Index v1Artificial Analysis290.7†264.4
HELM Capabilities meanHELM Capabilities79.9%80.7%†
Safety & honesty
FORTRESSScale AI SEAL41.7%17.0%†
MASKScale AI SEAL42.6%79.3%†
PropensityBenchScale AI SEAL52.9%–
Metadata
AA Openness IndexArtificial Analysis5.6†5.6

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • GPT-5 (medium): BenchLeader Index 58.2 → 59.0.
  • Gemini 3 Pro: 1 score revision.
  • GPT-5 (medium): 1 new result (Aider Polyglot).
  • GPT-5 (medium): 2 new results (Global-MMLU-Lite (AA), AIME 2025 (AA)); 1 score revision.
  • GPT-5 (medium): 1 score revision.
  • Gemini 3 Pro: 1 new result (Arena Search).
  • GPT-5 (medium): 1 score revision.
  • GPT-5 (medium): 1 score revision.
  • GPT-5 (medium): 1 new result (AA Progress Index v1).
  • Gemini 3 Pro: 1 score revision.

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

Gemini 3 Pro vs GPT-5: questions

Is Gemini 3 Pro better than GPT-5?
Gemini 3 Pro and GPT-5 (medium) are neck and neck on quality (60.0 vs 59.0, within the ±6.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
Gemini 3 Pro vs GPT-5: which is cheaper, and will that change?
GPT-5 is cheaper today: $3.44 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 Pro or GPT-5 better for coding?
They are level in coding (58 vs 57 on the category index). On Arena WebDev, the top shared coding benchmark, it is 1441 for Gemini 3 Pro against 1417 for GPT-5.
Is Gemini 3 Pro or GPT-5 better for agents and tool use?
Gemini 3 Pro scores higher in agentic tasks (62 vs 58 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 69.4% for Gemini 3 Pro against 49.6% for GPT-5.
Do independent benchmarks agree on Gemini 3 Pro vs GPT-5?
Across the 12 benchmarks both have, Gemini 3 Pro leads 12, GPT-5 leads 0. Every overall index that rates both puts Gemini 3 Pro ahead.
More questions (1)
When was Gemini 3 Pro released?
Gemini 3 Pro was released on 18 Nov 2025.