Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs DeepSeek V3.2 Exp

DeepSeek V3.2 leads DeepSeek V3.2 Exp on the Artificial Analysis Intelligence Index, 21.5 to 16.6, and DeepSeek V3.2 Exp costs 2.1x less per output token ($0.41 vs $0.88 per million). Updated October 21, 2026; 53 benchmarks have a result for both.

Replace DeepSeek V3.2 Replace DeepSeek V3.2 Exp
DeepSeek V3.2DeepSeek
Coding
30.0
Overall
37.9
Output $/M
$0.88
30
DeepSeek V3.2 ExpDeepSeek
Coding
No data
Overall
No data
Output $/M
$0.41
22
vs boards won of 53 shared, 1 tied

DeepSeek V3.2 or DeepSeek V3.2 Exp: which to pick

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; DeepSeek V3.2 Exp has none, so the two cannot be ranked on coding.

Pick DeepSeek V3.2 Exp for output tokens 2.1x cheaper ($0.41 vs $0.88 per million) and work measured by LMArena Text: medicine and healthcare or LMArena Text: business and finance, where it leads.

Model ADeepSeek V3.2DeepSeek30.0/1003 independent benchmarks$0.99 per solved taskModel BDeepSeek V3.2 ExpDeepSeekNot benchmarked1 independent benchmarkNo cost per solved task
Shared results
53
DeepSeek V3.2 only
22
DeepSeek V3.2 Exp only
6
Like-for-like categories
8 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

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
DeepSeek V3.2 30.0 from 3 benchmarks, DeepSeek V3.2 Exp not benchmarked.
Widest category
Agentic work, 56.7 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: DeepSeek V3.2 90.6% against DeepSeek V3.2 Exp 33.9%, 56.7 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.88 against DeepSeek V3.2 Exp $0.41 per million output tokens, 2.1x apart.
Context window
Both take 164K tokens.

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
CategoryDeepSeek V3.2BarsDeepSeek V3.2 ExpTies
Coding60
Intelligence20
Agentic work10
Reasoning50
Math40
Knowledge32
Web design151 tie
General815
All shared boards30221 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index60
Coding benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
Aider polyglotNot measured74.2%Only DeepSeek V3.2 Exp measured
SWE-bench Verified70.0%Not measuredOnly DeepSeek V3.2 measured
SWE-bench Multilingual59.0%Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev13621272DeepSeek V3.2
AA Coding Index44.2Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Vals AI)80.7%Not measuredOnly DeepSeek V3.2 measured
SWE-bench Verified (Vals AI)67.6%Not measuredOnly DeepSeek V3.2 measured
Vibe Code Bench (Vals AI)5.1%Not measuredOnly DeepSeek V3.2 measured
LiveCodeBench (Artificial Analysis run)86.2%78.9%DeepSeek V3.2
Terminal-Bench Hard (Artificial Analysis run)35.6%31.1%DeepSeek V3.2
Terminal-Bench39.6%Not measuredOnly DeepSeek V3.2 measured
WeirdMLNot measured39.5%Only DeepSeek V3.2 Exp measured
LMArena Text: coding14531438DeepSeek V3.2
LMArena Text: software and IT14491438DeepSeek V3.2
LMArena WebDev: React1350Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev: HTML13711281DeepSeek V3.2
Neither model has a result on the other 23 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
AA Intelligence Index21.516.6DeepSeek V3.2
Epoch Capabilities Index146.1144.9DeepSeek V3.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
τ²-Bench Telecom90.6%33.9%DeepSeek V3.2
APEX-Agents21.3%Not measuredOnly DeepSeek V3.2 measured
TheAgentCompanyNot measured42.9%Only DeepSeek V3.2 Exp measured
Vending-Bench 2$1,034Not measuredOnly DeepSeek V3.2 measured
Berkeley Function Calling Leaderboard (BFCL)Not measured56.7%Only DeepSeek V3.2 Exp measured
Neither model has a result on the other 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
GPQA Diamond80.3%Not measuredOnly DeepSeek V3.2 measured
GPQA Diamond (Artificial Analysis run)84.0%79.7%DeepSeek V3.2
Humanity's Last Exam (Artificial Analysis run)24.6%14.9%DeepSeek V3.2
ARC-AGI57.0%Not measuredOnly DeepSeek V3.2 measured
ARC-AGI-24.0%Not measuredOnly DeepSeek V3.2 measured
LMArena Text: hard prompts14341429DeepSeek V3.2
LMArena Text: expert14361422DeepSeek V3.2
LMArena Text: hard prompts in English14461444DeepSeek V3.2
Kagi LLM BenchmarkNot measured52.2%Only DeepSeek V3.2 Exp measured
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index40
Math benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
AIME84.6%Not measuredOnly DeepSeek V3.2 measured
MGSM (Vals AI)86.0%Not measuredOnly DeepSeek V3.2 measured
AA Math Index92.087.7DeepSeek V3.2
AIME 2025 (Artificial Analysis run)92.0%87.7%DeepSeek V3.2
LMArena Text: math14351422DeepSeek V3.2
LMArena Text: mathematical work14401433DeepSeek V3.2
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index32
Knowledge benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
MMLU Pro84.9%Not measuredOnly DeepSeek V3.2 measured
LegalBench (Vals AI)76.1%Not measuredOnly DeepSeek V3.2 measured
CaseLaw (Vals AI)55.4%Not measuredOnly DeepSeek V3.2 measured
CorpFin (Vals AI)51.0%Not measuredOnly DeepSeek V3.2 measured
TaxEval (Vals AI)68.2%Not measuredOnly DeepSeek V3.2 measured
MedQA (Vals AI)93.9%Not measuredOnly DeepSeek V3.2 measured
MMLU Pro (Artificial Analysis run)86.2%85.0%DeepSeek V3.2
LMArena Text: science14461443DeepSeek V3.2
LMArena Text: medicine and healthcare14401468DeepSeek V3.2 Exp
LMArena Text: legal and government14251419DeepSeek V3.2
LMArena Text: business and finance14101412DeepSeek V3.2 Exp
Neither model has a result on the other 7 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index15
Web design benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
Design Arena: websites11751179DeepSeek V3.2 Exp
Design Arena: UI components11491167DeepSeek V3.2 Exp
Design Arena: game dev11321144DeepSeek V3.2 Exp
Design Arena: data viz11581152DeepSeek V3.2
Design Arena: 3D11331148DeepSeek V3.2 Exp
Design Arena: all code categories11641171DeepSeek V3.2 Exp
Design Arena: SVG10321032Tie
Design Arena: ASCII art1083Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index815
General benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek V3.2 ExpLeader
IFBench60.7%54.1%DeepSeek V3.2
Long context reasoning (AA-LCR)73.3%72.3%DeepSeek V3.2
DTBench85.6%87.7%DeepSeek V3.2 Exp
LMCA28.8%29.1%DeepSeek V3.2 Exp
Fiction.LiveBenchNot measured83.3%Only DeepSeek V3.2 Exp measured
CL-bench12.4%13.2%DeepSeek V3.2 Exp
CL-bench Life7.4%9.5%DeepSeek V3.2 Exp
LMArena Text14251425DeepSeek V3.2 Exp
LMArena Text: instruction following14131401DeepSeek V3.2
LMArena Text: creative writing14001404DeepSeek V3.2 Exp
LMArena Text: English14391441DeepSeek V3.2 Exp
LMArena Text: non-English14091404DeepSeek V3.2
LMArena Text: Chinese14551461DeepSeek V3.2 Exp
LMArena Text: German14131440DeepSeek V3.2 Exp
LMArena Text: Japanese13741331DeepSeek V3.2
LMArena Text: Korean13701372DeepSeek V3.2 Exp
LMArena Text: French14291434DeepSeek V3.2 Exp
LMArena Text: Spanish14241441DeepSeek V3.2 Exp
LMArena Text: Russian14241412DeepSeek V3.2
LMArena Text: Polish14191435DeepSeek V3.2 Exp
LMArena Text: multi-turn14271422DeepSeek V3.2
LMArena Text: longer queries14281418DeepSeek V3.2
LMArena Text: writing and language14071407DeepSeek V3.2 Exp
LMArena Text: entertainment and media13951407DeepSeek V3.2 Exp
Neither model has a result on the other 22 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V3.2 and DeepSeek V3.2 Exp
BenchmarkDeepSeek V3.2DeepSeek 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

DeepSeek V3.2$0.0007Fits 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

DeepSeek V3.2$0.017Fits 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

DeepSeek V3.2$0.067Exceeds 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 DeepSeek V3.2 and DeepSeek V3.2 Exp
SpecificationDeepSeek V3.2DeepSeek V3.2 Exp
MakerDeepSeekDeepSeek
WeightsOpen weightsOpen weights
ReleasedDecember 1, 2025September 29, 2025
Context window164K164K
Input $/Mtok$0.29$0.27
Output $/Mtok$0.88$0.41
Coding index30.0Not benchmarked
Independent measurements behind it31
Cost per solved task$0.99Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 better than DeepSeek V3.2 Exp for coding?

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; DeepSeek V3.2 Exp has none, so the two cannot be ranked on coding. Pick DeepSeek V3.2 Exp for output tokens 2.1x cheaper ($0.41 vs $0.88 per million) and work measured by LMArena Text: medicine and healthcare or LMArena Text: business and finance, where it leads.

Which is cheaper, DeepSeek V3.2 or DeepSeek V3.2 Exp?

DeepSeek V3.2 costs $0.29 per million input tokens and $0.88 per million output tokens; DeepSeek V3.2 Exp costs $0.27 and $0.41. Cost per solved task is known for DeepSeek V3.2 only.

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

Neither: both take 164K tokens.

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

53 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.2's coding index rests on 3 independent benchmarks and DeepSeek V3.2 Exp's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next