BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases9

Gemini 3.7 Flash vs GPT-5.4 Pro

Full profiles: Gemini 3.7 Flash (Google) and GPT-5.4 Pro (OpenAI): every benchmark, price and rank.

Updated · Data from 10 sources

As of 25 Oct 2026, Gemini 3.7 Flash (high) and GPT-5.4 Pro are neck and neck on quality (63.5 vs 63.3 on the BenchLeader Index, inside the ±4.9 margin of error), and Gemini 3.7 Flash (high) costs 45× less, which makes it the better buy for most workloads. GPT-5.4 Pro leads in every category we measure both in.

Gemini 3.7 Flash benchmarks high Google GPT-5.4 Pro benchmarks OpenAIOpen in the comparison tool

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
ReasoningGPT-5.4 Pro73 vs 67 category scoreMedium
KnowledgeGPT-5.4 Pro74 vs 68 category scoreMedium
Images and other mediaGPT-5.4 Pro70 vs 68 category scoreMedium
Chat quality (human preference)–Only Gemini 3.7 Flash has an Arena rating (68 category score)Not comparable
Lowest costGemini 3.7 Flash$1.50 vs $67.50 per 1M blended; agentic coding session $0.060 vs $2.52–
Fastest first answerGemini 3.7 Flash2.5 s (Gemini 3.7 Flash (high)) vs 6.1 s (GPT-5.4 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 3.7 Flash (high) and GPT-5.4 Pro are neck and neck on quality (63.5 vs 63.3, within the ±4.9 margin).
  • Gemini 3.7 Flash (high) is stronger in agents & tools, coding, composite, human preference, long context, maths.
  • GPT-5.4 Pro is stronger in knowledge, multimodal, reasoning, instruction following.
  • Gemini 3.7 Flash (high) is 45× cheaper ($1.50 vs $67.50 per 1M blended).
  • Gemini 3.7 Flash (high) streams 12.6× faster (88 vs 7 tokens per second).

At a glance

SpecGemini 3.7 FlashGPT-5.4 Pro
DeveloperGoogleOpenAI
Released13 Aug 20264 Mar 2026
Context window (tokens)1,048,5761,050,000
Compared atGemini 3.7 Flash (high)GPT-5.4 Pro
Open weightsNoNo
Price in / out / cached input per 1M$0.750 / $3.75 / $0.075$30.00 / $180.00 / –

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 3.7 Flash (high)GPT-5.4 ProLeader
Reasoning67.372.5GPT-5.4 Pro
Coding62.9–not comparable
Agents & tools47.6–not comparable
Maths63.3–not comparable
Knowledge68.073.6GPT-5.4 Pro
Instruction following–63.8not comparable
Human preference68.4–not comparable
Multimodal67.669.8GPT-5.4 Pro
Long context65.9–not comparable
Composite73.2–not comparable

Where sources disagree

Across the 0 benchmarks both have, Gemini 3.7 Flash leads 0, GPT-5.4 Pro leads 0.

Overall indexGemini 3.7 Flash (high)GPT-5.4 ProLeader
BenchLeader Index63.563.3neck and neck

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; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.

MeasureGemini 3.7 Flash (high)GPT-5.4 Pro
Blended price per 1M tokens$1.50$67.50
Input / output per 1M$0.750 / $3.75$30.00 / $180.00
Cached input per 1M$0.075–
Price per task: chat reply$0.0014$0.066
Price per task: summarize a 30-page report$0.011$0.468
Price per task: code edit$0.010$0.450
Price per task: agentic coding session$0.060$2.52
Price per task: structured extraction$0.0023$0.096
Output speed88 tok/s7 tok/s
Time to first answer2.5 s6.1 s

Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.7 Flash (high) scores 63.5 and takes 2 s to its first answer token; Gemini 3.7 Flash (medium) scores 63.1 in 2 s, Gemini 3.7 Flash (low) scores 60.3 in 2 s. GPT-5.4 Pro scores 63.3 and takes 6 s to its first answer token; GPT-5.4 Pro (xhigh) scores 61.9 in 6 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

105 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 3.7 Flash (high)GPT-5.4 Pro
BenchLeader IndexBenchLeader63.563.3
Agents & tools scoreBenchLeader47.6–
Coding scoreBenchLeader62.9–
Composite scoreBenchLeader73.2–
Human preference scoreBenchLeader68.4–
Knowledge scoreBenchLeader68.073.6
Long context scoreBenchLeader65.9–
Maths scoreBenchLeader63.3–
Multimodal scoreBenchLeader67.669.8
Reasoning scoreBenchLeader67.372.5
Instruction following scoreBenchLeader–63.8
Blended price $/MBenchLeader$1.50$67.50
Output speedBenchLeader88 tok/s7 tok/s
Time to first answerBenchLeader2.5 s6.1 s
Context windowBenchLeader1,048,576 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub94.8%94.6%†
Humanity's Last ExamScale AI / CAIS–44.3%
SimpleBenchSimpleBench–74.1%
Arena Hard PromptsArena (LMArena)1509–
LiveBench 2.0 ReasoningLiveBench87.8%†–
LiveBench 1.0 ReasoningLiveBench87.8%–
GPQA Diamond (AA)Artificial Analysis94.5%–
Humanity's Last Exam (AA)Artificial Analysis47.9%–
GPQA Diamond (Vals)Vals AI93.9%–
ARC-AGI-1ARC Prize95.5%94.5%†
ARC-AGI-2ARC Prize84.6%83.3%†
CritPtArtificial Analysis14.3%30.0%†
Chess PuzzlesEpoch AI Benchmarking Hub66.0%59.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub37.0%–
LMCAEpoch AI Benchmarking Hub50.4%–
DTBenchEpoch AI Benchmarking Hub96.8%–
ForecastBenchForecastBench66.2%–
Coding
SciCodeSciCode57.2%–
WeirdMLWeirdML–57.4%†
FrontierCodeCognition43.6%†–
Arena CodingArena (LMArena)1520–
Arena WebDevArena (LMArena)1590–
LiveBench 2.0 CodingLiveBench80.5%†–
LiveBench 1.0 CodingLiveBench78.9%–
LiveCodeBenchVals AI88.7%–
Show all 105 benchmarks
MetricPublisherGemini 3.7 Flash (high)GPT-5.4 Pro
Coding
IOIVals AI67.8%–
SWE-bench (Vals)Vals AI80.8%–
Code MigrationVals AI34.8%–
ProgramBenchVals AI0.0%–
Vibe Code Bench v1.1Vals AI70.4%–
DeepSWE v1.1Epoch AI Benchmarking Hub65.3%–
ALE-BenchEpoch AI Benchmarking Hub904.3–
FrontierSWEEpoch AI Benchmarking Hub20.3%–
Terminal-Bench 4.0 (AA)Artificial Analysis13.6%–
Terminal-Bench 2.1 (AA)Artificial Analysis85.8%–
Agents & tools
Terminal-BenchTerminal-Bench11.2%–
Remote Labor IndexScale AI / CAIS5.0%†–
APEX-AgentsMercor67.8%†–
Arena AgentArena (LMArena)-1.3–
LiveBench Long AgenticLiveBench28.3%†–
LiveBench BrowserLiveBench56.0%†–
LiveBench Agentic CodingLiveBench58.3%–
Terminal-Bench 2.1 (Vals)Vals AI77.5%–
τ³-Banking (AA)Artificial Analysis32.8%–
Analyst Agent (AA)Artificial Analysis60.0%–
CyberBenchVals AI43.8%–
Finance Agent v2Vals AI59.0%–
Harvey's Legal Agent Benchmark v1Vals AI8.8%–
Legal Research BenchVals AI34.6%–
SkillsBenchVals AI65.9%–
SREBenchVals AI4.6%–
Tax Agent BenchVals AI22.8%–
Terminal-Bench 4.0 (Vals)Vals AI12.1%–
Terminal-Bench ScienceVals AI7.1%–
AutomationBenchArtificial Analysis62.0%–
GDP.pdf (AA)Artificial Analysis23.6%–
EnterpriseOps-GymArtificial Analysis50.4%†–
FACTS SearchGoogle80.9%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub71.6%82.5%†
FrontierMath Tier 4Epoch AI Benchmarking Hub36.6%58.5%†
OTIS Mock AIMEEpoch AI Benchmarking Hub97.2%–
ProofBenchVals AI58.0%†–
LiveBench MathematicsLiveBench93.5%–
MathArena ApexMathArena–69.8%†
Arena MathsArena (LMArena)1505–
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub69.2%46.3%†
LiveBench 2.0 Data AnalysisLiveBench68.0%†–
LiveBench ResearchLiveBench76.5%†–
LiveBench 1.0 Data AnalysisLiveBench68.0%–
AA-OmniscienceArtificial Analysis26.5–
MMLU-ProVals AI90.1%–
LegalBenchVals AI87.3%–
TaxEvalVals AI74.7%–
MultiNRCScale AI SEAL–62.3%
Excel Modeling BenchmarkVals AI71.3%–
MedCodeVals AI53.4%–
MedScribeVals AI83.9%–
GDP.pdfEpoch AI Benchmarking Hub23.8%–
AA-Omniscience: accuracyArtificial Analysis55.3%–
AA-Omniscience: non-hallucinationArtificial Analysis35.5%–
Instruction following
LiveBench LanguageLiveBench85.5%–
LiveBench Instruction FollowingLiveBench79.9%–
MultiChallengeScale AI SEAL–69.2%
TutorBenchScale AI SEAL–56.6%
Arena Instruction FollowingArena (LMArena)1487–
Human preference
Arena TextArena (LMArena)1490–
Arena Creative WritingArena (LMArena)1492–
Arena Multi-turnArena (LMArena)1489–
Arena Longer QueriesArena (LMArena)1501–
Multimodal
Arena VisionArena (LMArena)1314–
LiveBench DesignLiveBench73.2%†–
MMMU-ProArtificial Analysis85.5%–
VISTAScale AI SEAL–53.9%
MMMU-Pro (Vals)Vals AI89.0%–
MortgageTaxVals AI66.7%–
SAGEVals AI49.2%–
Long context
AA-LCRArtificial Analysis81.7%–
MLCRArtificial Analysis12.2%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub158.0†159.6
LiveBench 2.0LiveBench67.2%†–
LiveBench 1.0LiveBench78.8%–
AA Intelligence Index v4.3.2Artificial Analysis39.1–
Vals RSI IndexVals AI18.3%–
Safety & honesty
FORTRESSScale AI SEAL–14.8%
MASKScale AI SEAL–91.7%

Data as of 2026-10-25. Best configuration of each model; every score links to its source on the model pages.

What changed

  • Gemini 3.7 Flash (high): 2 new results (Harvey's Legal Agent Benchmark v1, Vals RSI Index).
  • Gemini 3.7 Flash (high): input price moved from $0.38/M to $0.75/M (97% up); output price moved from $1.88/M to $3.75/M (99% up).
  • Gemini 3.7 Flash (high): 1 score revision; input price moved from $0.75/M to $0.38/M (49% down); output price moved from $3.75/M to $1.88/M (50% down).
  • GPT-5.4 Pro: 1 score revision.
  • Gemini 3.7 Flash (high): 1 new result (ForecastBench).
  • Gemini 3.7 Flash (high): BenchLeader Index 62.9 → 63.5.
  • Gemini 3.7 Flash (high): BenchLeader Index 60.8 → 62.9.
  • Gemini 3.7 Flash (high): 18 new results (AA Intelligence Index v4.3.2, GPQA Diamond (AA), Humanity's Last Exam (AA)…).
  • Gemini 3.7 Flash (high): BenchLeader Index 63.2 → 60.8.
  • Gemini 3.7 Flash (high): 2 score revisions.

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

Gemini 3.7 Flash vs GPT-5.4 Pro: questions

Is Gemini 3.7 Flash better than GPT-5.4 Pro?
Gemini 3.7 Flash (high) and GPT-5.4 Pro are neck and neck on quality (63.5 vs 63.3, within the ±4.9 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2026-10-25, so check the category scores for your use.
Which is cheaper, Gemini 3.7 Flash or GPT-5.4 Pro, and will that change?
Gemini 3.7 Flash is cheaper today: $1.50 against $67.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Which is better for long documents?
Both take about 1.1M tokens: Gemini 3.7 Flash 1,048,576, GPT-5.4 Pro 1,050,000.
Which has lower latency?
Gemini 3.7 Flash (high) answers first: 2.5 s against 6.1 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.
Which is faster, Gemini 3.7 Flash or GPT-5.4 Pro?
Gemini 3.7 Flash streams faster: 88 against 7 output tokens per second.
When was Gemini 3.7 Flash released?
Gemini 3.7 Flash was released on 13 Aug 2026.