LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Claude Opus 4.1 vs DeepSeek V3.2 Exp

Updated October 21, 2026. 42 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Claude Opus 4.1 Replace DeepSeek V3.2 Exp
Claude Opus 4.1Anthropic
Coding
27.2
Overall
22.8
Output $/M
$75.00
27
DeepSeek V3.2 ExpDeepSeek
Coding
55.0
Overall
16.6
Output $/M
$0.41
14
vs boards won of 42 shared, 1 tied

Decision reading

Pick DeepSeek V3.2 Exp for coding: it scores 55.0 on the Vibe Coding Index against 27.2 for Claude Opus 4.1.

Pick Claude Opus 4.1 when the context window is larger, 200K against 164K tokens, or when what matters is LMArena WebDev or WeirdML, where it leads.

Model AClaude Opus 4.1Anthropic27.2/1001 independent benchmarkNo cost per solved taskModel BDeepSeek V3.2 ExpDeepSeek55.0/1002 independent benchmarksNo cost per solved task
Shared results
42
Claude Opus 4.1 only
29
DeepSeek V3.2 Exp only
15
Like-for-like categories
7 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
Claude Opus 4.1 27.2 from 1 benchmark, DeepSeek V3.2 Exp 55.0 from 2 benchmarks.
Widest category
Coding, 10.5 points apart on average over 5 shared benchmarks.
Widest single gap
LMArena WebDev: HTML: Claude Opus 4.1 1399 against DeepSeek V3.2 Exp 1281, 19.6 points apart on a 0 to 100 scale.
Output price
Claude Opus 4.1 $75.00 against DeepSeek V3.2 Exp $0.41 per million output tokens, 182.9x apart.
Context window
Claude Opus 4.1 takes 200K tokens, DeepSeek V3.2 Exp 164K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryClaude Opus 4.1BarsDeepSeek V3.2 ExpTies
Coding50
Intelligence11
Reasoning30
Math20
Knowledge13
Web design421 tie
General118
All shared boards27141 tie
CodingLike-for-likeFour of these boards feed the Vibe Coding Index50
Coding benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
Aider polyglotNot measured74.2%Only DeepSeek V3.2 Exp measured
LMArena WebDev13901272Claude Opus 4.1
LiveCodeBench (Vals AI)66.5%Not measuredOnly Claude Opus 4.1 measured
LiveCodeBench (Artificial Analysis run)Not measured78.9%Only DeepSeek V3.2 Exp measured
Terminal-Bench Hard (Artificial Analysis run)Not measured31.1%Only DeepSeek V3.2 Exp measured
Terminal-Bench38.0%Not measuredOnly Claude Opus 4.1 measured
WeirdML45.9%39.5%Claude Opus 4.1
Cybench42.0%Not measuredOnly Claude Opus 4.1 measured
LMArena Text: coding14791438Claude Opus 4.1
LMArena Text: software and IT14561438Claude Opus 4.1
LMArena WebDev: HTML13991281Claude Opus 4.1
Neither model has a result on the other 27 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
AA Intelligence Index22.816.6Claude Opus 4.1
Epoch Capabilities Index144.1144.9DeepSeek V3.2 Exp
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
Agentic work benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
τ²-Bench TelecomNot measured33.9%Only DeepSeek V3.2 Exp measured
METR time horizons66.8%Not measuredOnly Claude Opus 4.1 measured
GDPval43.6%Not measuredOnly Claude Opus 4.1 measured
DeepResearch Bench48.3%Not measuredOnly Claude Opus 4.1 measured
TheAgentCompanyNot measured42.9%Only DeepSeek V3.2 Exp measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index30
Reasoning benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
GPQA Diamond76.3%Not measuredOnly Claude Opus 4.1 measured
MMMU Pro (Vals AI)77.5%Not measuredOnly Claude Opus 4.1 measured
GPQA Diamond (Artificial Analysis run)Not measured79.7%Only DeepSeek V3.2 Exp measured
Humanity's Last Exam (Artificial Analysis run)Not measured14.9%Only DeepSeek V3.2 Exp measured
GPQA Diamond (Epoch AI run)77.3%Not measuredOnly Claude Opus 4.1 measured
SimpleBench60.0%Not measuredOnly Claude Opus 4.1 measured
Mystery game puzzles21.0%Not measuredOnly Claude Opus 4.1 measured
EBR-bench7.9%Not measuredOnly Claude Opus 4.1 measured
VPCT35.0%Not measuredOnly Claude Opus 4.1 measured
LMArena Text: hard prompts14421429Claude Opus 4.1
LMArena Text: expert14381422Claude Opus 4.1
LMArena Text: hard prompts in English14511444Claude Opus 4.1
Neither model has a result on the other 9 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
AIME78.2%Not measuredOnly Claude Opus 4.1 measured
MATH 500 (Vals AI)95.4%Not measuredOnly Claude Opus 4.1 measured
MGSM (Vals AI)94.4%Not measuredOnly Claude Opus 4.1 measured
AA Math IndexNot measured87.7Only DeepSeek V3.2 Exp measured
AIME 2025 (Artificial Analysis run)Not measured87.7%Only DeepSeek V3.2 Exp measured
OTIS Mock AIME68.9%Not measuredOnly Claude Opus 4.1 measured
FrontierMath, tiers 1 to 312.6%Not measuredOnly Claude Opus 4.1 measured
FrontierMath, tier 42.4%Not measuredOnly Claude Opus 4.1 measured
LMArena Text: math14311422Claude Opus 4.1
LMArena Text: mathematical work14341433Claude Opus 4.1
Neither model has a result on the other 3 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index13
Knowledge benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
MMLU Pro87.9%Not measuredOnly Claude Opus 4.1 measured
LegalBench (Vals AI)83.5%Not measuredOnly Claude Opus 4.1 measured
TaxEval (Vals AI)73.7%Not measuredOnly Claude Opus 4.1 measured
MortgageTax (Vals AI)61.1%Not measuredOnly Claude Opus 4.1 measured
MedQA (Vals AI)93.6%Not measuredOnly Claude Opus 4.1 measured
MedCode (Vals AI)47.2%Not measuredOnly Claude Opus 4.1 measured
MMLU Pro (Artificial Analysis run)Not measured85.0%Only DeepSeek V3.2 Exp measured
LMArena Text: science14221443DeepSeek V3.2 Exp
LMArena Text: medicine and healthcare14161468DeepSeek V3.2 Exp
LMArena Text: legal and government14211419Claude Opus 4.1
LMArena Text: business and finance14061412DeepSeek V3.2 Exp
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index42
Web design benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
Design Arena: websites11771179DeepSeek V3.2 Exp
Design Arena: UI components11641166DeepSeek V3.2 Exp
Design Arena: game dev11721145Claude Opus 4.1
Design Arena: data viz11661152Claude Opus 4.1
Design Arena: 3D11531148Claude Opus 4.1
Design Arena: all code categories11711171Tie
Design Arena: SVG11521031Claude Opus 4.1
Design Arena: ASCII art1167Not measuredOnly Claude Opus 4.1 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index118
General benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
SAGE (Vals AI)31.4%Not measuredOnly Claude Opus 4.1 measured
MedScribe (Vals AI)73.9%Not measuredOnly Claude Opus 4.1 measured
IFBenchNot measured54.1%Only DeepSeek V3.2 Exp measured
Long context reasoning (AA-LCR)Not measured72.3%Only DeepSeek V3.2 Exp measured
DTBench80.0%87.7%DeepSeek V3.2 Exp
LMCA37.1%29.1%Claude Opus 4.1
Fiction.LiveBenchNot measured83.3%Only DeepSeek V3.2 Exp measured
Creative writing (Lech Mazur)84.7%Not measuredOnly Claude Opus 4.1 measured
CL-benchNot measured13.2%Only DeepSeek V3.2 Exp measured
CL-bench LifeNot measured9.5%Only DeepSeek V3.2 Exp measured
LMArena Text14191425DeepSeek V3.2 Exp
LMArena Text: instruction following14351401Claude Opus 4.1
LMArena Text: creative writing14121404Claude Opus 4.1
LMArena Text: English14331441DeepSeek V3.2 Exp
LMArena Text: non-English14051404Claude Opus 4.1
LMArena Text: Chinese14271461DeepSeek V3.2 Exp
LMArena Text: German14131440DeepSeek V3.2 Exp
LMArena Text: Japanese13771331Claude Opus 4.1
LMArena Text: Korean13811372Claude Opus 4.1
LMArena Text: French14331434DeepSeek V3.2 Exp
LMArena Text: Spanish14481441Claude Opus 4.1
LMArena Text: Russian14221412Claude Opus 4.1
LMArena Text: Polish14131435DeepSeek V3.2 Exp
LMArena Text: multi-turn14441422Claude Opus 4.1
LMArena Text: longer queries14551418Claude Opus 4.1
LMArena Text: writing and language14181407Claude Opus 4.1
LMArena Text: entertainment and media13981407DeepSeek V3.2 Exp
Neither model has a result on the other 19 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Claude Opus 4.1 and DeepSeek V3.2 Exp
BenchmarkClaude Opus 4.1DeepSeek V3.2 ExpLeader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

Claude Opus 4.1$0.052Fits in one request
DeepSeek V3.2 Exp$0.0005Fits in one request

DeepSeek V3.2 Exp costs less

Repository review

50K input + 3K output tokens

Claude Opus 4.1$0.975Fits in one request
DeepSeek V3.2 Exp$0.015Fits in one request

DeepSeek V3.2 Exp costs less

Agent loop

200K input + 10K output tokens

Claude Opus 4.1$3.75Exceeds the context window
DeepSeek V3.2 Exp$0.058Exceeds the context window

DeepSeek V3.2 Exp costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of Claude Opus 4.1 and DeepSeek V3.2 Exp
SpecificationAnthropic Claude Opus 4.1DeepSeek V3.2 Exp
MakerAnthropicDeepSeek
WeightsNo public weights listedOpen weights
ReleasedAugust 5, 2025September 29, 2025
Context window200K164K
Input $/Mtok$15.00$0.27
Output $/Mtok$75.00$0.41
Coding index27.255.0
Independent measurements behind it12
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Claude Opus 4.1 better than DeepSeek V3.2 Exp for coding?

Pick DeepSeek V3.2 Exp for coding: it scores 55.0 on the Vibe Coding Index against 27.2 for Claude Opus 4.1. Pick Claude Opus 4.1 when the context window is larger, 200K against 164K tokens, or when what matters is LMArena WebDev or WeirdML, where it leads.

Which is cheaper, Claude Opus 4.1 or DeepSeek V3.2 Exp?

Claude Opus 4.1 costs $15.00 per million input tokens and $75.00 per million output tokens; DeepSeek V3.2 Exp costs $0.27 and $0.41. Cost per solved task is known for neither of them.

Which has the bigger context window, Claude Opus 4.1 or DeepSeek V3.2 Exp?

Claude Opus 4.1, with 200K tokens against 164K.

How many benchmarks is the Claude Opus 4.1 vs DeepSeek V3.2 Exp comparison based on?

42 benchmarks have a result for both models as of October 21, 2026. Claude Opus 4.1's coding index rests on 1 independent benchmark and DeepSeek V3.2 Exp's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next