BenchLeader

DeepSeek V4 Pro vs GPT-6 Sol

Full profiles: DeepSeek V4 Pro (DeepSeek) and GPT-6 Sol (OpenAI): every benchmark, price and rank.

Updated · Data from 12 sources

As of 16 Nov 2026, GPT-6 Sol is ahead of DeepSeek V4 Pro on quality (62.6 vs 56.0 on the BenchLeader Index), but DeepSeek V4 Pro is 4.0× cheaper at $0.990 vs $4.00 per 1M tokens blended. GPT-6 Sol leads in every category we measure both in. GPT-6 Sol streams 3×+ faster.

Key facts (as of 16 Nov 2026)

  • DeepSeek V4 Pro scores 56.0 on the BenchLeader Index and GPT-6 Sol scores 62.6 (BenchLeader, 16 Nov 2026).
  • DeepSeek V4 Pro costs $0.990 per 1M tokens blended (list price, 16 Nov 2026).
  • GPT-6 Sol costs $4.00 per 1M tokens blended (list price, 16 Nov 2026).
  • GPT-6 Sol streams 3×+ faster than DeepSeek V4 Pro (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 33 benchmarks both models have, GPT-6 Sol leads 28 (BenchLeader, 16 Nov 2026).
  • On Arena WebDev, DeepSeek V4 Pro is rated 1581 and GPT-6 Sol 1673 (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
CodingGPT-6 Sol66 vs 52 category score; Arena WebDev 1673 vs 1581High
MathsGPT-6 Sol68 vs 65 category score; OTIS Mock AIME 100.0% vs 95.6%Medium
ReasoningGPT-6 Sol66 vs 56 category score; GPQA Diamond 94.3% vs 90.9%High
KnowledgeGPT-6 Sol61 vs 41 category score; AA-Omniscience v1.1 21.4 vs -6.5High
Long documentsGPT-6 Sol67 vs 64 category score; AA-LCR v1.1 83.7% vs 79.3%Medium
Chat quality (human preference)Either65 vs 64 on Arena votesEither
Lowest costDeepSeek V4 Pro$1.98 vs $4.00 per 1M blended; agentic coding session $0.095 vs $0.160–
Fastest first answerEither42.8 s (DeepSeek V4 Pro (high)) vs 36.9 s (GPT-6 Sol (max)) 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

  • GPT-6 Sol (max) leads on quality: 62.6 vs 56.0.
  • DeepSeek V4 Pro (high) is stronger in human preference.
  • GPT-6 Sol (max) is stronger in coding, knowledge, long context, maths, reasoning.
  • DeepSeek V4 Pro (high) is 4.0× cheaper ($0.990 vs $4.00 per 1M blended).
  • GPT-6 Sol (max) streams 1.6× faster (127 vs 80 tokens per second).

At a glance

SpecDeepSeek V4 ProGPT-6 Sol
DeveloperDeepSeekOpenAI
Released13 Aug 202622 Sep 2026
Context window (tokens)1,048,5761,050,000
Compared atDeepSeek V4 Pro (high)GPT-6 Sol (max)
Open weightsYesNo
Price in / out / cached input per 1M$1.32 / $3.96 / $0.022$2.00 / $10.00 / $0.200

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 Pro (high)GPT-6 Sol (max)Leader
Reasoning55.666.0GPT-6 Sol
Coding51.566.3GPT-6 Sol
Agents & tools–58.8not comparable
Maths64.568.0GPT-6 Sol
Knowledge41.260.6GPT-6 Sol
Human preference65.064.2level
Multimodal–60.6not comparable
Long context64.366.6GPT-6 Sol

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 avg75ReasoningCodingMathsKnowledgeLong contextHumanpreferenceDeepSeek V4 Pro, Reasoning: 55.6DeepSeek V4 Pro, Coding: 51.5DeepSeek V4 Pro, Maths: 64.5DeepSeek V4 Pro, Knowledge: 41.2DeepSeek V4 Pro, Long context: 64.3DeepSeek V4 Pro, Human preference: 65.0GPT-6 Sol, Reasoning: 66.0GPT-6 Sol, Coding: 66.3GPT-6 Sol, Maths: 68.0GPT-6 Sol, Knowledge: 60.6GPT-6 Sol, Long context: 66.6GPT-6 Sol, Human preference: 64.2

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

Model, click to add or removeIndex

Where sources disagree

Across the 33 benchmarks both have, DeepSeek V4 Pro leads 5, GPT-6 Sol leads 28. Independent sources split: DeepSeek V4 Pro leads on Arena Text; GPT-6 Sol leads on BenchLeader Index and AA Intelligence Index v5.0.1.

Overall indexDeepSeek V4 Pro (high)GPT-6 Sol (max)Leader
BenchLeader Index56.062.6GPT-6 Sol
AA Intelligence Index v5.0.120.844.2GPT-6 Sol
Arena Text14641457DeepSeek V4 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 Pro (high)GPT-6 Sol (max)
Blended price per 1M tokens$1.98$4.00
Input / output per 1M$1.32 / $3.96$2.00 / $10.00
Cached input per 1M$0.022$0.200
Price per task: chat reply$0.0017$0.0038
Price per task: summarize a 30-page report$0.018$0.030
Price per task: code edit$0.014$0.027
Price per task: agentic coding session$0.095$0.160
Price per task: structured extraction$0.0034$0.0060
Agentic coding session, input mostly cached$0.037$0.079
Cost per benchmark task (AA Index)$0.849$2.47
Output tokens per task (AA Index)70k54k
Of which reasoning45k35k
Time per task (AA Index)15 min7 min
Output speed80 tok/s127 tok/s
Time to first answer42.8 s36.9 s
Time to a complete answer, last measured 13 Nov 2026–143.8 s

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $0.849 on DeepSeek V4 Pro (high) and $2.47 on GPT-6 Sol (max), and DeepSeek V4 Pro (high) wrote 1.3× as many output tokens for the same work, which a per-token price hides.

Each model is compared at its best-ranked reasoning-effort setting. DeepSeek V4 Pro (high) scores 56.0 and takes 43 s to its first answer token; DeepSeek V4 Pro (max) scores 54.7 in 47 s, DeepSeek V4 Pro scores 53.0 in 1 s. GPT-6 Sol (max) scores 62.6 and takes 37 s to its first answer token; GPT-6 Sol (xhigh) scores 58.4 in 19 s, GPT-6 Sol (high) scores 58.1 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

159 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 Pro (high)GPT-6 Sol (max)
BenchLeader IndexBenchLeader56.062.6
Coding scoreBenchLeader51.566.3
Human preference scoreBenchLeader65.064.2
Knowledge scoreBenchLeader41.260.6
Long context scoreBenchLeader64.366.6
Maths scoreBenchLeader64.568.0
Reasoning scoreBenchLeader55.666.0
Agents & tools scoreBenchLeader–58.8
Multimodal scoreBenchLeader–60.6
Blended price $/MBenchLeader$1.98$4.00
Output speedBenchLeader80 tok/s127 tok/s
Time to first answerBenchLeader42.8 s36.9 s
Context windowBenchLeader1,048,576 tokens1,050,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub90.9%94.3%
SimpleBenchSimpleBench–73.1%†
Arena Hard PromptsArena (LMArena)14821483
LiveBench 1.0 ReasoningLiveBench85.8%†88.7%
GPQA Diamond (AA)Artificial Analysis92.8%†–
Humanity's Last Exam (AA)Artificial Analysis41.5%47.9%
HLE-Diamond (AA)Artificial Analysis20.5%45.2%
GPQA Diamond (Vals)Vals AI92.4%†–
Kagi LLM BenchmarkKagi LLM Benchmark53.5%†–
ARC-AGI-1ARC Prize87.2%95.5%
ARC-AGI-2ARC Prize59.7%89.6%
ARC-AGI-3ARC Prize–23.0%
CritPtArtificial Analysis18.6%30.9%
CritPt v2 hard subset (AA)Artificial Analysis28.5%66.5%
MysteryMechanismVals AI–30.2%
HLE-Diamond (Scale)Scale AI SEAL13.4%†33.8%†
Arena ExpertArena (LMArena)14811496
Chess PuzzlesEpoch AI Benchmarking Hub16.0%56.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub43.0%†56.0%
CL-bench LifeEpoch AI Benchmarking Hub13.5%–
LMCAEpoch AI Benchmarking Hub45.5%†59.1%
DTBenchEpoch AI Benchmarking Hub93.9%†97.3%
ForecastBenchForecastBench60.7%†–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub77.6%†–
SciCodeSciCode46.4%57.6%
Show all 159 benchmarks
MetricPublisherDeepSeek V4 Pro (high)GPT-6 Sol (max)
Coding
WeirdMLWeirdML46.5%–
FrontierCodeCognition28.6%49.3%
Arena CodingArena (LMArena)15031521
Arena WebDevArena (LMArena)15811673
LiveBench 1.0 CodingLiveBench77.2%†81.8%
SciCode (AA)Artificial Analysis51.7%57.6%
LiveCodeBenchVals AI87.5%†–
IOIVals AI51.6%†82.6%
SWE-bench (Vals)Vals AI96.4%†–
Code MigrationVals AI41.5%†57.2%
Code Migration v1.1Vals AI52.0%†–
ProgramBenchVals AI0.0%†2.0%
Vibe Code Bench 1-100Vals AI17.5%†–
Vibe Code Bench v1.1Vals AI82.3%†87.8%
SWE Atlas v1.0: Codebase QnAScale AI SEAL27.1%†–
SWE Atlas v1.0: Test WritingScale AI SEAL27.1%†–
MirrorCodeEpoch AI Benchmarking Hub33.3%†48.9%
Surface Evolver BenchEpoch AI Benchmarking Hub40.0%–
ALE-BenchEpoch AI Benchmarking Hub1006.12462
WeirdML v3WeirdML–19.7%†
Terminal-Bench 4.0 (AA)Artificial Analysis6.1%43.9%
Terminal-Bench 2.1 (AA)Artificial Analysis78.7%†–
Startrise FrontendStartrise62.7%†–
GameBench 2 OverallDuelLab GameBench 241.9†73.0†
GameBench 2 by reasoning settingDuelLab GameBench 245.179.4†
Design Arena WebsiteDesign Arena1227†1290†
FrontierCode 1.1 (AA)Artificial Analysis38.8%†–
FrontierCode 2.0 (AA)Artificial Analysis16.7%35.6%
Agents & tools
Terminal-BenchTerminal-Bench6.7%†50.3%
APEX-AgentsMercor47.3%†54.3%
Arena AgentArena (LMArena)–6.9
LiveBench Agentic CodingLiveBench55.0%†52.9%
Terminal-Bench 2.1 (Vals)Vals AI50.2%†83.2%
HiL-BenchScale AI SEAL38.5%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis47.5%†–
GDPval-AA v2.1Artificial Analysis13361514
τ³-Banking (AA)Artificial Analysis39.6%†–
ITBench SRE (AA)Artificial Analysis–49.4%
CyberBenchVals AI72.9%†78.0%
Finance Agent v2Vals AI50.4%†49.0%
Finance Agent v2.1Vals AI57.1%†–
Harvey's Legal Agent Benchmark v1.1Vals AI10.8%†–
Harvey's Legal Agent Benchmark v1Vals AI7.5%†1.7%
Legal Research BenchVals AI40.9%†28.9%
SkillsBenchVals AI53.8%†–
SREBenchVals AI1.9%†30.5%
Tax Agent BenchVals AI26.8%†15.0%
Terminal-Bench 2.0 (Vals)Vals AI56.2%†–
Terminal-Bench 4.0 (Vals)Vals AI14.1%†44.4%
Terminal-Bench ScienceVals AI4.3%†30.0%
Arena SearchArena (LMArena)–1229
Arena Agent Task CompletionArena (LMArena)–4.5
Vending-Bench 2Andon Labs3284.5†14427.9†
SpireClimbEpoch AI Benchmarking Hub–9.2%
AutomationBenchArtificial Analysis46.2%61.6%
GDP.pdf (AA)Artificial Analysis10.4%25.2%
Harvey LAB-AA v1.1Harvey–3.6%
EnterpriseOps-GymArtificial Analysis49.6%†–
AA-Briefcase v1.1Artificial Analysis11481476
Design Arena AgenticDesign Arena927†1218†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis1.9%30.0%
CWE-Bench (AA)Artificial Analysis–64.2%
DeepSec-Bench (AA)Artificial Analysis–46.3%
CyberGym end-to-end (AA)Artificial Analysis–0.0%
FACTS SearchGoogle–92.7%†
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub64.6%†89.8%
FrontierMath Tier 4Epoch AI Benchmarking Hub26.8%†90.0%
OTIS Mock AIMEEpoch AI Benchmarking Hub95.6%100.0%
ProofBenchVals AI50.0%†83.0%†
LiveBench MathematicsLiveBench95.1%†96.4%
AIME 2026MathArena96.7%†–
HMMT February 2026MathArena93.9%†–
USAMO 2026MathArena60.7%†–
MathArena ApexMathArena28.1%†–
Arena MathsArena (LMArena)14651440
FrontierMath ErdősEpoch AI Benchmarking Hub–1.5%
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub52.9%†60.7%
LiveBench 1.0 Data AnalysisLiveBench79.2%†81.2%
AA-Omniscience v1.1Artificial Analysis-6.521.4
MMLU-ProVals AI87.3%†–
LegalBenchVals AI82.4%†–
CorpFinVals AI65.4%†–
TaxEvalVals AI73.1%†–
PRBench LegalScale AI SEAL–38.6%†
BioMysteryBenchVals AI–74.8%
CaseLaw v2Vals AI59.4%†–
Excel Modeling BenchmarkVals AI52.8%†71.5%
Excel Modeling Benchmark v1.1Vals AI65.1%†–
MedCodeVals AI42.5%†47.1%
MedScribeVals AI80.2%†82.0%
Public Benefits Bench v1.1Vals AI62.9%†56.6%
Financial Coaching BenchVals AI–62.6%
Immigration BenchVals AI61.7%†–
DrugDiscoveryBenchScale AI SEAL31.7%†–
EBR-benchEpoch AI Benchmarking Hub–53.3%
GDP.pdfEpoch AI Benchmarking Hub–26.4%
AA-Omniscience v1.1: accuracyArtificial Analysis45.1%51.1%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis5.9%39.1%
HealthBench Professional (AA)OpenAI–62.3%
FACTS ParametricGoogle–76.1%†
Instruction following
LiveBench LanguageLiveBench82.1%†85.3%
LiveBench Instruction FollowingLiveBench67.7%†68.6%
Arena Instruction FollowingArena (LMArena)14611449
Human preference
Arena TextArena (LMArena)14641457
EQ-Bench 4EQ-Bench1156†–
EQ-Bench Creative Writing v3EQ-Bench1553†2126†
Arena Creative WritingArena (LMArena)14441428
Arena Multi-turnArena (LMArena)14691468
Arena Longer QueriesArena (LMArena)14741466
Multimodal
Arena VisionArena (LMArena)–1233
MMMU-ProArtificial Analysis–83.0%
Humanity's Sixth SenseScale AI SEAL–31.2%
Blueprint-Bench 2Epoch AI Benchmarking Hub–36.9%†
Furniture AssemblyEpoch AI Benchmarking Hub–58.3%
AA-3DArtificial Analysis–1131
FACTS MultimodalGoogle–43.9%†
Long context
AA-LCR v1.1Artificial Analysis79.3%83.7%
MLCRArtificial Analysis–33.9%
FACTS GroundingGoogle–83.3%†
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub155.9†163.8†
LiveBench 1.0LiveBench77.4%†79.3%
AA Intelligence Index v5.0.1Artificial Analysis20.844.2
AA Intelligence Index v4.3.2Artificial Analysis36†47.6
AA Progress Index v1Artificial Analysis311.5†334
Vals IndexVals AI53.0†–
Vals Finance IndexVals AI52.9%†–
Vals RSI IndexVals AI–28.1%
Vals Science IndexVals AI–45.0%
Vals Router IndexVals AI53.0%†–
Vals Legal IndexVals AI69.0%†–
AA Cyber IndexArtificial Analysis–36.8
Safety & honesty
FORTRESSScale AI SEAL–13.5%†
DistressBenchScale AI SEAL–68.0%†
Metadata
AA Openness IndexArtificial Analysis44.4–

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-6 Sol (max): 1 new result (GDPval-AA v2.1).
  • DeepSeek V4 Pro (high): BenchLeader Index 59.1 → 56.0.
  • DeepSeek V4 Pro (high): 17 new results (AA Intelligence Index v5.0.1, Humanity's Last Exam (AA), HLE-Diamond (AA)…); input price moved from $0.66/M to $1.32/M (100% up); output price moved from $1.98/M to $3.96/M (100% up).
  • GPT-6 Sol (max): 1 new result (MirrorCode).
  • DeepSeek V4 Pro (high): BenchLeader Index 57.5 → 59.1.
  • GPT-6 Sol (max): BenchLeader Index 63.5 → 62.9.
  • GPT-6 Sol (max): 2 score revisions.
  • GPT-6 Sol (max): 1 score revision.
  • DeepSeek V4 Pro (high): 1 score revision.
  • GPT-6 Sol (max): 1 score revision.

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

DeepSeek V4 Pro vs GPT-6 Sol: questions

Is GPT-6 Sol better than DeepSeek V4 Pro?
GPT-6 Sol (max) leads on quality: 62.6 vs 56.0. The BenchLeader Index combines every independent quality benchmark; GPT-6 Sol (max) is ahead overall as of 16 Nov 2026, but check the category scores for your use.
DeepSeek V4 Pro vs GPT-6 Sol: which is cheaper, and will that change?
DeepSeek V4 Pro is cheaper today: $0.990 against $4.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is DeepSeek V4 Pro or GPT-6 Sol better for coding?
GPT-6 Sol scores higher in coding (66 vs 52 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1581 for DeepSeek V4 Pro against 1673 for GPT-6 Sol.
Which is faster, DeepSeek V4 Pro or GPT-6 Sol?
GPT-6 Sol streams 3×+ faster (127 vs 80 output tokens per second, Artificial Analysis 7-day median).
Which is better for long documents?
GPT-6 Sol scores higher on long-context benchmarks (67 vs 64). Both take about 1.1M tokens: DeepSeek V4 Pro 1,048,576, GPT-6 Sol 1,050,000.
More questions (3)
Which has lower latency?
GPT-6 Sol answers first: 36.9 s against 42.8 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 DeepSeek V4 Pro vs GPT-6 Sol?
Across the 33 benchmarks both have, DeepSeek V4 Pro leads 5, GPT-6 Sol leads 28. Independent sources split: DeepSeek V4 Pro leads on Arena Text; GPT-6 Sol leads on BenchLeader Index and AA Intelligence Index v5.0.1.
When was GPT-6 Sol released?
GPT-6 Sol was released on 22 Sep 2026.