BenchLeader

DeepSeek V4 Flash vs Gemini 3 Pro

Full profiles: DeepSeek V4 Flash (DeepSeek) and Gemini 3 Pro (Google): every benchmark, price and rank.

Updated · Data from 15 sources

As of 16 Nov 2026, DeepSeek V4 Flash and Gemini 3 Pro are neck and neck on quality (57.2 vs 60.0 on the BenchLeader Index, inside the ±5.2 margin), and DeepSeek V4 Flash is 17× cheaper at $0.262 vs $4.50 per 1M tokens blended. Choose DeepSeek V4 Flash for maths (62 vs 54); choose Gemini 3 Pro for human preference (68 vs 62), reasoning (63 vs 59), and coding (58 vs 55).

Key facts (as of 16 Nov 2026)

  • DeepSeek V4 Flash scores 57.2 on the BenchLeader Index and Gemini 3 Pro scores 60.0 (BenchLeader, 16 Nov 2026).
  • DeepSeek V4 Flash costs $0.262 per 1M tokens blended (list price, 16 Nov 2026).
  • Gemini 3 Pro costs $4.50 per 1M tokens blended (list price, 16 Nov 2026).
  • Of the 15 benchmarks both models have, Gemini 3 Pro leads 11 (BenchLeader, 16 Nov 2026).
  • On Arena WebDev, DeepSeek V4 Flash is rated 1582 and Gemini 3 Pro 1441 (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
CodingGemini 3 Pro58 vs 55 category score; WeirdML 69.9% vs 57.0%Medium
MathsDeepSeek V4 Flash62 vs 54 category score; Arena Maths 1441 vs 1475High
ReasoningGemini 3 Pro63 vs 59 category score; Arena Hard Prompts 1502 vs 1459Medium
KnowledgeGemini 3 Pro56 vs 55 category score; MMLU-Pro 90.1% vs 86.2%Medium
Chat quality (human preference)Gemini 3 Pro68 vs 62 on Arena votesMedium
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • DeepSeek V4 Flash (high) and Gemini 3 Pro are neck and neck on quality (57.2 vs 60.0, within the ±5.2 margin).
  • DeepSeek V4 Flash (high) is stronger in maths.
  • Gemini 3 Pro is stronger in coding, human preference, knowledge, reasoning.
  • DeepSeek V4 Flash (high) is 17× cheaper ($0.262 vs $4.50 per 1M blended).

At a glance

SpecDeepSeek V4 FlashGemini 3 Pro
DeveloperDeepSeekGoogle
Released31 Jul 202618 Nov 2025
Context window (tokens)1,048,576–
Compared atDeepSeek V4 Flash (high)Gemini 3 Pro
Open weightsYesNo
Price in / out / cached input per 1M$0.150 / $0.600 / $0.0030–

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.

CategoryDeepSeek V4 Flash (high)Gemini 3 ProLeader
Reasoning59.262.6Gemini 3 Pro
Coding55.057.8Gemini 3 Pro
Agents & tools–61.7not comparable
Maths61.954.0DeepSeek V4 Flash
Knowledge54.855.9Gemini 3 Pro
Instruction following–61.1not comparable
Human preference62.067.7Gemini 3 Pro
Multimodal–65.6not 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 avg75ReasoningCodingMathsKnowledgeHumanpreferenceDeepSeek V4 Flash, Reasoning: 59.2DeepSeek V4 Flash, Coding: 55.0DeepSeek V4 Flash, Maths: 61.9DeepSeek V4 Flash, Knowledge: 54.8DeepSeek V4 Flash, Human preference: 62.0Gemini 3 Pro, Reasoning: 62.6Gemini 3 Pro, Coding: 57.8Gemini 3 Pro, Maths: 54.0Gemini 3 Pro, Knowledge: 55.9Gemini 3 Pro, Human preference: 67.7

Not drawn, measured for fewer than two of these models: Agents & tools, Instruction following, Long context, Multimodal.

Model, click to add or removeIndex

Where sources disagree

Across the 15 benchmarks both have, DeepSeek V4 Flash leads 4, Gemini 3 Pro leads 11. Every overall index that rates both puts Gemini 3 Pro ahead.

Overall indexDeepSeek V4 Flash (high)Gemini 3 ProLeader
BenchLeader Index57.260.0neck and neck
Arena Text14391486Gemini 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.

MeasureDeepSeek V4 Flash (high)Gemini 3 Pro
Blended price per 1M tokens$0.262–
Input / output per 1M$0.150 / $0.600– / –
Cached input per 1M$0.0030–
Price per task: chat reply$0.0002–
Price per task: summarize a 30-page report$0.0022–
Price per task: code edit$0.0018–
Price per task: agentic coding session$0.011–
Price per task: structured extraction$0.0004–
Agentic coding session, input mostly cached$0.0048–
Output speed82 tok/s–
Time to first answer0.7 s–

Each model is compared at its best-ranked reasoning-effort setting. DeepSeek V4 Flash (high) scores 57.2 and takes 1 s to its first answer token; DeepSeek V4 Flash (max) scores 56.8 in 28 s, DeepSeek V4 Flash scores 53.8 in 1 s. Gemini 3 Pro scores 60.0; Gemini 3 Pro (high) scores 60.0. 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

147 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.

MetricPublisherDeepSeek V4 Flash (high)Gemini 3 Pro
BenchLeader IndexBenchLeader57.260.0
Coding scoreBenchLeader55.057.8
Human preference scoreBenchLeader62.067.7
Knowledge scoreBenchLeader54.855.9
Maths scoreBenchLeader61.954.0
Reasoning scoreBenchLeader59.262.6
Agents & tools scoreBenchLeader–61.7
Instruction following scoreBenchLeader–61.1
Multimodal scoreBenchLeader–65.6
Blended price $/MBenchLeader$0.262–
Output speedBenchLeader82 tok/s–
Time to first answerBenchLeader0.7 s–
Context windowBenchLeader1,048,576 tokens–
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub91.0%†92.6%
Humanity's Last ExamScale AI / CAIS–37.5%
SimpleBenchSimpleBench61.1%†76.4%
Arena Hard PromptsArena (LMArena)14591502
LiveBench 1.0 ReasoningLiveBench86.6%†–
GPQA Diamond (AA)Artificial Analysis90.8%†90.8%†
Humanity's Last Exam (AA)Artificial Analysis38.5%†39.7%†
GPQA Diamond (Vals)Vals AI89.9%91.7%†
Kagi LLM BenchmarkKagi LLM Benchmark52.2%†80.1%
GPQA Diamond (HELM)HELM Capabilities–80.3%
ARC-AGI-1ARC Prize87.0%75.0%
ARC-AGI-2ARC Prize56.0%31.1%
CritPtArtificial Analysis16.6%†9.1%†
SciPredictScale AI SEAL–25.3%
Arena ExpertArena (LMArena)14671499
Chess PuzzlesEpoch AI Benchmarking Hub32.0%†–
Mystery Game PuzzlesEpoch AI Benchmarking Hub34.0%†–
CL-benchEpoch AI Benchmarking Hub–15.8%
LMCAEpoch AI Benchmarking Hub41.7%†–
DTBenchEpoch AI Benchmarking Hub90.9%†–
ForecastBenchForecastBench–65.0%
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–72.9%
SciCodeSciCode42.0%–
WeirdMLWeirdML57.0%69.9%
FrontierCodeCognition18.8%†–
Show all 147 benchmarks
MetricPublisherDeepSeek V4 Flash (high)Gemini 3 Pro
Coding
GSO-BenchGSO-Bench–18.6%
Arena CodingArena (LMArena)14801517
Arena WebDevArena (LMArena)15821441
LiveBench 1.0 CodingLiveBench75.0%†–
SciCode (AA)Artificial Analysis50.4%†–
LiveCodeBenchVals AI87.3%86.4%
IOIVals AI32.7%–
SWE-bench (Vals)Vals AI88.8%76.4%†
SWE-Bench ProScale AI SEAL–43.3%
SWE-bench Verified (bash only)SWE-bench–74.2%
SWE-bench Multilingual (bash only)SWE-bench–68.7%
SWE-bench Verified (any scaffold)SWE-bench–77.4%
Code MigrationVals AI38.6%–
ProgramBenchVals AI0.0%–
Vibe Code Bench v1.1Vals AI74.7%14.3%†
SWE-Bench Pro (private)Scale AI SEAL–17.9%
ALE-BenchEpoch AI Benchmarking Hub678.21176.8
AlgoTuneAlgoTune–1.8
Terminal-Bench 4.0 (AA)Artificial Analysis12.1%†–
Terminal-Bench 2.1 (AA)Artificial Analysis78.7%†–
GameBench 2 OverallDuelLab GameBench 234.1†–
GameBench 2 by reasoning settingDuelLab GameBench 243.4–
Design Arena WebsiteDesign Arena1199†1244
LiveCodeBench (AA)Artificial Analysis–91.8%†
Agents & tools
Terminal-BenchTerminal-Bench–69.4%
GDPvalOpenAI–40.3%
Remote Labor IndexScale AI / CAIS–1.3%
LiveBench Agentic CodingLiveBench46.8%†–
Terminal-Bench HardArtificial Analysis–41.7%†
τ²-Bench Telecom (AA)Artificial Analysis–87.1%†
Terminal-Bench 2.1 (Vals)Vals AI67.0%†–
MCP AtlasScale AI SEAL–70.3%
HiL-BenchScale AI SEAL47.8%†–
BFCL OverallBerkeley Function Calling Leaderboard–72.5%
τ²-benchτ²-bench–82.5%†
τ³-bench Bankingτ²-bench–18.0%†
GDPval-AA v2.1Artificial Analysis1437†–
τ³-Banking (AA)Artificial Analysis39.4%†–
CyberBenchVals AI72.1%–
Finance Agent v2Vals AI49.5%–
Harvey's Legal Agent Benchmark v1Vals AI8.3%–
Legal Research BenchVals AI30.3%–
Poker AgentVals AI–1078.9%
SkillsBenchVals AI50.7%–
Tax Agent BenchVals AI28.1%–
Terminal-Bench 2.0 (Vals)Vals AI–55.1%†
Terminal-Bench 4.0 (Vals)Vals AI18.7%–
Terminal-Bench ScienceVals AI4.3%–
Arena SearchArena (LMArena)–1201
BALROGBALROG–58.1%
DeepResearch BenchDeepResearch Bench–46.3%†
METR Time HorizonsMETR–71.0%
Vending-Bench 2Andon Labs–5478.2
AutomationBenchArtificial Analysis54.0%†–
GDP.pdf (AA)Artificial Analysis11.0%†–
AA-Briefcase v1.1Artificial Analysis1255†–
Design Arena AgenticDesign Arena–1080
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub57.5%†–
FrontierMath Tier 4Epoch AI Benchmarking Hub24.4%†–
OTIS Mock AIMEEpoch AI Benchmarking Hub94.4%†91.4%
ProofBenchVals AI56.0%†20.0%
LiveBench MathematicsLiveBench86.8%†–
AIME (Vals)Vals AI–96.7%†
MGSMVals AI–93.9%†
AIME 2026MathArena95.8%†91.7%
HMMT February 2026MathArena93.9%†86.4%
MathArena ApexMathArena27.1%†23.4%
Omni-MATH (HELM)HELM Capabilities–55.6%
MATH 500Vals AI–96.4%†
Arena MathsArena (LMArena)14411475
AIME 2025 (AA)Artificial Analysis–95.7%†
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub33.6%†–
LiveBench 1.0 Data AnalysisLiveBench79.3%†–
MMLU-ProVals AI86.2%90.1%
LegalBenchVals AI77.7%87.0%†
CorpFinVals AI61.9%63.7%†
TaxEvalVals AI70.7%72.6%†
MedQAVals AI–96.0%†
PRBench FinanceScale AI SEAL–39.2%
PRBench LegalScale AI SEAL–40.6%
MultiNRCScale AI SEAL–59.0%
MMLU-Pro (HELM)HELM Capabilities–90.3%
BioMysteryBenchVals AI64.4%–
CaseLaw v2Vals AI–53.1%†
Excel Modeling BenchmarkVals AI57.0%–
MedCodeVals AI41.4%52.2%†
MedScribeVals AI80.4%72.0%†
Global-MMLU-Lite (AA)Artificial Analysis–92.2%†
Instruction following
LiveBench LanguageLiveBench79.2%†–
LiveBench Instruction FollowingLiveBench65.5%†–
IFBenchArtificial Analysis–70.4%†
MultiChallengeScale AI SEAL–65.7%
TutorBenchScale AI SEAL–53.7%
IFEval (HELM)HELM Capabilities–87.6%
WildBench (HELM)HELM Capabilities–85.9%
Arena Instruction FollowingArena (LMArena)14341473
Human preference
Arena TextArena (LMArena)14391486
EQ-Bench Creative Writing v3EQ-Bench1559†1524
Arena Creative WritingArena (LMArena)14111484
Arena Multi-turnArena (LMArena)14471495
Arena Longer QueriesArena (LMArena)14491491
Multimodal
Arena VisionArena (LMArena)–1305
MMMU-ProArtificial Analysis–80.2%†
VISTAScale AI SEAL–51.5%
MMMU-Pro (Vals)Vals AI–87.5%†
MortgageTaxVals AI–69.1%†
AudioMCScale AI SEAL–54.6%†
AudioMC (text output)Scale AI SEAL–54.6%†
VTBScale AI SEAL–26.9%
GeoBenchGeoBench–84.0%
VPCTEpoch AI Benchmarking Hub–91.0%
Long context
AA-LCR v1.1Artificial Analysis79.7%†76.0%†
Arena DocumentArena (LMArena)–1451
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub154.4†153.0
LiveBench 1.0LiveBench74.2%†–
AA Intelligence Index v4.3.2Artificial Analysis34.3†28.0†
AA Progress Index v1Artificial Analysis309.4†290.7†
HELM Capabilities meanHELM Capabilities–79.9%
Safety & honesty
FORTRESSScale AI SEAL–41.7%
MASKScale AI SEAL–42.6%
PropensityBenchScale AI SEAL–52.9%
Metadata
AA Openness IndexArtificial Analysis44.4†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

  • Gemini 3 Pro: 1 score revision.
  • DeepSeek V4 Flash (high): 1 score revision.
  • Gemini 3 Pro: 1 new result (Arena Search).
  • DeepSeek V4 Flash (high): 1 score revision.
  • DeepSeek V4 Flash (high): 1 new result (GameBench 2 by reasoning setting).
  • DeepSeek V4 Flash (high): 1 new result (Harvey's Legal Agent Benchmark v1).
  • Gemini 3 Pro: 1 score revision.
  • DeepSeek V4 Flash (high): BenchLeader Index 59.3 → 57.3.
  • DeepSeek V4 Flash (high): input price moved from $0.11/M to $0.15/M (36% up); output price moved from $0.23/M to $0.60/M (161% up).
  • Gemini 3 Pro: 1 score revision.

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

DeepSeek V4 Flash vs Gemini 3 Pro: questions

Is DeepSeek V4 Flash better than Gemini 3 Pro?
DeepSeek V4 Flash (high) and Gemini 3 Pro are neck and neck on quality (57.2 vs 60.0, within the ±5.2 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.
DeepSeek V4 Flash vs Gemini 3 Pro: which is cheaper, and will that change?
DeepSeek V4 Flash is cheaper today: $0.262 against $4.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is DeepSeek V4 Flash or Gemini 3 Pro better for coding?
Gemini 3 Pro scores higher in coding (58 vs 55 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1582 for DeepSeek V4 Flash against 1441 for Gemini 3 Pro.
Do independent benchmarks agree on DeepSeek V4 Flash vs Gemini 3 Pro?
Across the 15 benchmarks both have, DeepSeek V4 Flash leads 4, Gemini 3 Pro leads 11. Every overall index that rates both puts Gemini 3 Pro ahead.
When was DeepSeek V4 Flash released?
DeepSeek V4 Flash was released on 31 Jul 2026.