BenchLeader

Claude Opus 4.8 vs DeepSeek V4 Pro

Full profiles: Claude Opus 4.8 (Anthropic) and DeepSeek V4 Pro (DeepSeek): every benchmark, price and rank.

Updated · Data from 14 sources

As of 16 Nov 2026, Claude Opus 4.8 and DeepSeek V4 Pro are neck and neck on quality (61.1 vs 56.0 on the BenchLeader Index, inside the ±5.9 margin), and DeepSeek V4 Pro is 10× cheaper at $0.990 vs $10.00 per 1M tokens blended. Claude Opus 4.8 leads in every category we measure both in. Claude Opus 4.8 streams ~1.5× faster.

Key facts (as of 16 Nov 2026)

  • Claude Opus 4.8 scores 61.1 on the BenchLeader Index and DeepSeek V4 Pro scores 56.0 (BenchLeader, 16 Nov 2026).
  • Claude Opus 4.8 costs $10.00 per 1M tokens blended (list price, 16 Nov 2026).
  • DeepSeek V4 Pro costs $0.990 per 1M tokens blended (list price, 16 Nov 2026).
  • Claude Opus 4.8 streams ~1.5× faster than DeepSeek V4 Pro (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 14 benchmarks both models have, Claude Opus 4.8 leads 12 (BenchLeader, 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
CodingClaude Opus 4.862 vs 52 category score; SciCode 53.5% vs 46.4%High
MathsClaude Opus 4.868 vs 65 category score; OTIS Mock AIME 98.3% vs 95.6%Medium
ReasoningClaude Opus 4.864 vs 56 category score; GPQA Diamond 91.0% vs 90.9%High
KnowledgeClaude Opus 4.861 vs 41 category scoreHigh
Long documentsEither63 vs 64 category score; AA-LCR v1.1 77.7% vs 79.3%Either
Chat quality (human preference)–Only DeepSeek V4 Pro has an Arena rating (65 category score)Not comparable
Lowest costDeepSeek V4 Pro$1.98 vs $10.00 per 1M blended; agentic coding session $0.095 vs $0.400–
Fastest first answerEither39.1 s (Claude Opus 4.8 (max)) vs 42.8 s (DeepSeek V4 Pro (high)) 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

  • Claude Opus 4.8 (max) and DeepSeek V4 Pro (high) are neck and neck on quality (61.1 vs 56.0, within the ±5.9 margin).
  • Claude Opus 4.8 (max) is stronger in coding, knowledge, maths, reasoning.
  • DeepSeek V4 Pro (high) is stronger in long context.
  • DeepSeek V4 Pro (high) is 10× cheaper ($0.990 vs $10.00 per 1M blended).

At a glance

SpecClaude Opus 4.8DeepSeek V4 Pro
DeveloperAnthropicDeepSeek
Released28 May 202613 Aug 2026
Context window (tokens)1,000,0001,048,576
Compared atClaude Opus 4.8 (max)DeepSeek V4 Pro (high)
Open weightsNoYes
Price in / out / cached input per 1M$5.00 / $25.00 / $0.500$1.32 / $3.96 / $0.022

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.

CategoryClaude Opus 4.8 (max)DeepSeek V4 Pro (high)Leader
Reasoning64.155.6Claude Opus 4.8
Coding61.951.5Claude Opus 4.8
Agents & tools56.9–not comparable
Maths67.764.5Claude Opus 4.8
Knowledge61.041.2Claude Opus 4.8
Instruction following62.1–not comparable
Human preference–65.0not comparable
Long context63.464.3level

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 contextClaude Opus 4.8, Reasoning: 64.1Claude Opus 4.8, Coding: 61.9Claude Opus 4.8, Maths: 67.7Claude Opus 4.8, Knowledge: 61.0Claude Opus 4.8, Long context: 63.4DeepSeek V4 Pro, Reasoning: 55.6DeepSeek V4 Pro, Coding: 51.5DeepSeek V4 Pro, Maths: 64.5DeepSeek V4 Pro, Knowledge: 41.2DeepSeek V4 Pro, Long context: 64.3

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

Model, click to add or removeIndex

Where sources disagree

Across the 14 benchmarks both have, Claude Opus 4.8 leads 12, DeepSeek V4 Pro leads 2. Every overall index that rates both puts Claude Opus 4.8 ahead.

Overall indexClaude Opus 4.8 (max)DeepSeek V4 Pro (high)Leader
BenchLeader Index61.156.0neck and neck
Arena Text1482†1464Claude Opus 4.8

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.

MeasureClaude Opus 4.8 (max)DeepSeek V4 Pro (high)
Blended price per 1M tokens$10.00$1.98
Input / output per 1M$5.00 / $25.00$1.32 / $3.96
Cached input per 1M$0.500$0.022
Price per task: chat reply$0.0095$0.0017
Price per task: summarize a 30-page report$0.075$0.018
Price per task: code edit$0.068$0.014
Price per task: agentic coding session$0.400$0.095
Price per task: structured extraction$0.015$0.0034
Agentic coding session, input mostly cached$0.198$0.037
Cost per benchmark task (AA Index)–$0.849
Output tokens per task (AA Index)–70k
Of which reasoning–45k
Time per task (AA Index)–15 min
Output speed85 tok/s80 tok/s
Time to first answer39.1 s42.8 s
Time to a complete answer, last measured 13 Nov 202681.0 s–

Each model is compared at its best-ranked reasoning-effort setting. Claude Opus 4.8 (max) scores 61.1 and takes 39 s to its first answer token; Claude Opus 4.8 (high) scores 60.4 in 1 s, Claude Opus 4.8 scores 58.7 in 1 s. 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. 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

158 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.

MetricPublisherClaude Opus 4.8 (max)DeepSeek V4 Pro (high)
BenchLeader IndexBenchLeader61.156.0
Agents & tools scoreBenchLeader56.9–
Coding scoreBenchLeader61.951.5
Instruction following scoreBenchLeader62.1–
Knowledge scoreBenchLeader61.041.2
Long context scoreBenchLeader63.464.3
Maths scoreBenchLeader67.764.5
Reasoning scoreBenchLeader64.155.6
Human preference scoreBenchLeader–65.0
Blended price $/MBenchLeader$10.00$1.98
Output speedBenchLeader85 tok/s80 tok/s
Time to first answerBenchLeader39.1 s42.8 s
Context windowBenchLeader1,000,000 tokens1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub91.0%90.9%
SimpleBenchSimpleBench64.8%†–
Arena Hard PromptsArena (LMArena)1505†1482
LiveBench 1.0 ReasoningLiveBench89.2%85.8%†
GPQA Diamond (AA)Artificial Analysis92.0%92.8%†
Humanity's Last Exam (AA)Artificial Analysis48.7%41.5%
HLE-Diamond (AA)Artificial Analysis–20.5%
GPQA Diamond (Vals)Vals AI92.4%92.4%†
EnigmaEvalScale AI SEAL23.5%†–
Kagi LLM BenchmarkKagi LLM Benchmark88.8%†53.5%†
ARC-AGI-1ARC Prize92.5%87.2%
ARC-AGI-2ARC Prize72.1%†59.7%
ARC-AGI-3ARC Prize1.5%†–
CritPtArtificial Analysis20.9%18.6%
CritPt v2 hard subset (AA)Artificial Analysis–28.5%
HLE-Diamond (Scale)Scale AI SEAL–13.4%†
Arena ExpertArena (LMArena)1514†1481
Chess PuzzlesEpoch AI Benchmarking Hub52.0%16.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub36.0%43.0%†
CL-bench LifeEpoch AI Benchmarking Hub–13.5%
LMCAEpoch AI Benchmarking Hub57.5%45.5%†
DTBenchEpoch AI Benchmarking Hub94.9%93.9%†
ForecastBenchForecastBench66.1%60.7%†
BTF-3Epoch AI Benchmarking Hub14.0%†–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–77.6%†
Show all 158 benchmarks
MetricPublisherClaude Opus 4.8 (max)DeepSeek V4 Pro (high)
Coding
SciCodeSciCode53.5%46.4%
WeirdMLWeirdML70.5%†46.5%
FrontierCodeCognition46.5%†28.6%
GSO-BenchGSO-Bench47.1%†–
Arena CodingArena (LMArena)1529†1503
Arena WebDevArena (LMArena)1536†1581
LiveBench 1.0 CodingLiveBench81.8%77.2%†
SciCode (AA)Artificial Analysis54.4%51.7%
LiveCodeBenchVals AI87.8%87.5%†
IOIVals AI–51.6%†
SWE-bench (Vals)Vals AI88.6%96.4%†
Code MigrationVals AI47.3%41.5%†
Code Migration v1.1Vals AI–52.0%†
ProgramBenchVals AI1.0%0.0%†
Vibe Code Bench 1-100Vals AI–17.5%†
Vibe Code Bench v1.1Vals AI82.7%82.3%†
SWE Atlas v1.0: Codebase QnAScale AI SEAL57.3%†27.1%†
SWE Atlas v1.0: RefactoringScale AI SEAL46.7%†–
SWE Atlas v1.0: Test WritingScale AI SEAL49.6%†27.1%†
MirrorCodeEpoch AI Benchmarking Hub65.6%33.3%†
Surface Evolver BenchEpoch AI Benchmarking Hub87.5%†40.0%
DeepSWE v1.1Epoch AI Benchmarking Hub59.0%–
ALE-BenchEpoch AI Benchmarking Hub1563.8†1006.1
Terminal-Bench 4.0 (AA)Artificial Analysis21.7%6.1%
Terminal-Bench 2.1 (AA)Artificial Analysis84.6%78.7%†
Startrise FrontendStartrise–62.7%†
GameBench 2 OverallDuelLab GameBench 249.4†41.9†
GameBench 2 by reasoning settingDuelLab GameBench 245.1†45.1
Design Arena WebsiteDesign Arena1250†1227†
FrontierCode 1.1 (AA)Artificial Analysis–38.8%†
FrontierCode 2.0 (AA)Artificial Analysis–16.7%
Agents & tools
Terminal-BenchTerminal-Bench23.6%6.7%†
OSWorld-Verified 2.0OSWorld20.6%–
Remote Labor IndexScale AI / CAIS8.3%†–
APEX-AgentsMercor48.9%47.3%†
LiveBench Agentic CodingLiveBench50.5%55.0%†
Terminal-Bench HardArtificial Analysis58.3%–
τ²-Bench Telecom (AA)Artificial Analysis94.4%–
Terminal-Bench 2.1 (Vals)Vals AI71.9%50.2%†
MCP AtlasScale AI SEAL82.2%–
HiL-BenchScale AI SEAL35.3%†38.5%†
τ³-bench Bankingτ²-bench39.7%–
GDPval-AA v2.1, normalised scoreArtificial Analysis–47.5%†
GDPval-AA v2.1Artificial Analysis14611336
τ³-Banking (AA)Artificial Analysis34.2%39.6%†
Analyst Agent (AA)Artificial Analysis45.0%–
CyberBenchVals AI–72.9%†
Finance Agent v2Vals AI53.9%50.4%†
Finance Agent v2.1Vals AI–57.1%†
Harvey's Legal Agent Benchmark v1.1Vals AI–10.8%†
Harvey's Legal Agent Benchmark v1Vals AI9.6%7.5%†
Legal Research BenchVals AI43.8%40.9%†
SkillsBenchVals AI59.2%53.8%†
SREBenchVals AI–1.9%†
Tax Agent BenchVals AI30.0%26.8%†
Terminal-Bench 2.0 (Vals)Vals AI70.0%56.2%†
Terminal-Bench 4.0 (Vals)Vals AI23.2%14.1%†
Terminal-Bench ScienceVals AI4.3%4.3%†
Time Horizon Index: KSPVals AI11.8%–
Arena SearchArena (LMArena)1194†–
DeepResearch BenchDeepResearch Bench50.2%†–
PostTrainBenchEpoch AI Benchmarking Hub31.0%–
GBAEvalEpoch AI Benchmarking Hub70.9%†–
Vending-Bench 2Andon Labs2992.33284.5†
AutomationBenchArtificial Analysis45.6%46.2%
GDP.pdf (AA)Artificial Analysis22.8%10.4%
EnterpriseOps-GymArtificial Analysis44.0%49.6%†
AA-Briefcase v1.1Artificial Analysis13201148
Design Arena AgenticDesign Arena1226†927†
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–1.9%
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub80.0%64.6%†
FrontierMath Tier 4Epoch AI Benchmarking Hub56.1%26.8%†
OTIS Mock AIMEEpoch AI Benchmarking Hub98.3%95.6%
ProofBenchVals AI69.0%50.0%†
LiveBench MathematicsLiveBench94.3%95.1%†
AIME 2026MathArena100.0%96.7%†
HMMT February 2026MathArena95.5%93.9%†
USAMO 2026MathArena–60.7%†
MathArena ApexMathArena81.3%28.1%†
Arena MathsArena (LMArena)1476†1465
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub53.0%52.9%†
LiveBench 1.0 Data AnalysisLiveBench66.0%79.2%†
AA-Omniscience v1.1Artificial Analysis–-6.5
MMLU-ProVals AI89.6%87.3%†
LegalBenchVals AI83.6%82.4%†
CorpFinVals AI66.7%65.4%†
TaxEvalVals AI75.6%73.1%†
CaseLaw v2Vals AI–59.4%†
Excel Modeling BenchmarkVals AI69.4%52.8%†
Excel Modeling Benchmark v1.1Vals AI–65.1%†
MedCodeVals AI53.2%42.5%†
MedScribeVals AI85.8%80.2%†
Public Benefits Bench v1.1Vals AI68.1%62.9%†
Immigration BenchVals AI–61.7%†
DrugDiscoveryBenchScale AI SEAL46.8%31.7%†
EBR-benchEpoch AI Benchmarking Hub28.6%–
GDP.pdfEpoch AI Benchmarking Hub24.0%–
AA-Omniscience v1.1: accuracyArtificial Analysis–45.1%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–5.9%
Instruction following
LiveBench LanguageLiveBench79.7%82.1%†
LiveBench Instruction FollowingLiveBench72.0%67.7%†
IFBenchArtificial Analysis62.2%–
Arena Instruction FollowingArena (LMArena)1479†1461
Human preference
Arena TextArena (LMArena)1474†1464
EQ-Bench 4EQ-Bench1268†1156†
EQ-Bench Creative Writing v3EQ-Bench1840†1553†
Arena Creative WritingArena (LMArena)1464†1444
Arena Multi-turnArena (LMArena)1493†1469
Arena Longer QueriesArena (LMArena)1499†1474
Multimodal
Arena VisionArena (LMArena)1289†–
MMMU-Pro (Vals)Vals AI86.6%–
MortgageTaxVals AI69.9%–
Humanity's Sixth SenseScale AI SEAL23.5%–
GeoBenchGeoBench81.0%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub14.5%†–
Furniture AssemblyEpoch AI Benchmarking Hub42.5%–
Long context
AA-LCR v1.1Artificial Analysis77.7%79.3%
Arena DocumentArena (LMArena)1476†–
MLCRArtificial Analysis19.4%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub159.9†155.9†
LiveBench 1.0LiveBench76.2%77.4%†
AA Intelligence Index v5.0.1Artificial Analysis–20.8
AA Intelligence Index v4.3.2Artificial Analysis41.836†
AA Progress Index v1Artificial Analysis318.4311.5†
Vals IndexVals AI–53.0†
Vals Finance IndexVals AI–52.9%†
Vals RSI IndexVals AI24.0%–
Vals Multimodal IndexVals AI70.9%–
Vals Router IndexVals AI–53.0%†
Vals Legal IndexVals AI–69.0%†
Safety & honesty
FORTRESSScale AI SEAL18.2%–
DistressBenchScale AI SEAL71.4%†–
Metadata
AA Openness IndexArtificial Analysis–44.4

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

  • Claude Opus 4.8 (max): BenchLeader Index 60.3 → 61.1.
  • Claude Opus 4.8 (max): 27 new results (LiveCodeBench, MMLU-Pro, LegalBench…).
  • Claude Opus 4.8 (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).
  • Claude Opus 4.8 (max): 1 score revision.
  • Claude Opus 4.8 (max): 1 new result (MirrorCode).
  • DeepSeek V4 Pro (high): BenchLeader Index 57.5 → 59.1.
  • DeepSeek V4 Pro (high): 1 score revision.
  • DeepSeek V4 Pro (high): 1 score revision.

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

Claude Opus 4.8 vs DeepSeek V4 Pro: questions

Is Claude Opus 4.8 better than DeepSeek V4 Pro?
Claude Opus 4.8 (max) and DeepSeek V4 Pro (high) are neck and neck on quality (61.1 vs 56.0, within the ±5.9 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.
Claude Opus 4.8 vs DeepSeek V4 Pro: which is cheaper, and will that change?
DeepSeek V4 Pro is cheaper today: $0.990 against $10.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Opus 4.8 or DeepSeek V4 Pro better for coding?
Claude Opus 4.8 scores higher in coding (62 vs 52 on the category index, where 50 is average). On SciCode, the top shared coding benchmark, it is 53.5% for Claude Opus 4.8 against 46.4% for DeepSeek V4 Pro.
Which is faster, Claude Opus 4.8 or DeepSeek V4 Pro?
Claude Opus 4.8 streams ~1.5× faster (85 vs 80 output tokens per second, Artificial Analysis 7-day median).
Which is better for long documents?
They are level on long-context benchmarks (63 vs 64). Both take about 1M tokens: Claude Opus 4.8 1,000,000, DeepSeek V4 Pro 1,048,576.
More questions (3)
Which has lower latency?
Claude Opus 4.8 answers first: 39.1 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 Claude Opus 4.8 vs DeepSeek V4 Pro?
Across the 14 benchmarks both have, Claude Opus 4.8 leads 12, DeepSeek V4 Pro leads 2. Every overall index that rates both puts Claude Opus 4.8 ahead.
When was DeepSeek V4 Pro released?
DeepSeek V4 Pro was released on 13 Aug 2026.