LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 Exp vs Qwen3.5-35B-A3B

Updated October 21, 2026. 40 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace DeepSeek V3.2 Exp Replace Qwen3.5-35B-A3B
DeepSeek V3.2 ExpDeepSeek
Coding
55.0
Overall
16.6
Output $/M
$0.41
33
Qwen3.5-35B-A3BQwen
Coding
7.7
Overall
19.3
Output $/M
$1.00
7
vs boards won of 40 shared

Decision reading

Pick DeepSeek V3.2 Exp for coding: it scores 55.0 on the Vibe Coding Index against 7.7 for Qwen3.5-35B-A3B.

Pick Qwen3.5-35B-A3B when the context window is larger, 262K against 164K tokens, or when what matters is LMArena WebDev: HTML or AA Intelligence Index, where it leads.

Model ADeepSeek V3.2 ExpDeepSeek55.0/1002 independent benchmarksNo cost per solved taskModel BQwen3.5-35B-A3BQwen7.7/1001 independent benchmarkNo cost per solved task
Shared results
40
DeepSeek V3.2 Exp only
17
Qwen3.5-35B-A3B only
5
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
DeepSeek V3.2 Exp 55.0 from 2 benchmarks, Qwen3.5-35B-A3B 7.7 from 1 benchmark.
Widest category
Agentic work, 55.3 points apart on average over 1 shared benchmark.
Widest single gap
τ²-Bench Telecom: DeepSeek V3.2 Exp 33.9% against Qwen3.5-35B-A3B 89.2%, 55.3 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 Exp $0.41 against Qwen3.5-35B-A3B $1.00 per million output tokens, 2.4x apart.
Context window
DeepSeek V3.2 Exp takes 164K tokens, Qwen3.5-35B-A3B 262K.

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 V3.2 ExpBarsQwen3.5-35B-A3BTies
Coding41
Intelligence11
Agentic work01
Reasoning32
Math20
Knowledge40
General192
All shared boards337
CodingLike-for-likeFour of these boards feed the Vibe Coding Index41
Coding benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
Aider polyglot74.2%Not measuredOnly DeepSeek V3.2 Exp measured
LMArena WebDev12721254DeepSeek V3.2 Exp
LiveCodeBench (Artificial Analysis run)78.9%Not measuredOnly DeepSeek V3.2 Exp measured
Terminal-Bench Hard (Artificial Analysis run)31.1%26.5%DeepSeek V3.2 Exp
WeirdML39.5%Not measuredOnly DeepSeek V3.2 Exp measured
ALE-BenchNot measured221.8Only Qwen3.5-35B-A3B measured
LMArena Text: coding14381410DeepSeek V3.2 Exp
LMArena Text: software and IT14381415DeepSeek V3.2 Exp
LMArena WebDev: ReactNot measured1238Only Qwen3.5-35B-A3B measured
LMArena WebDev: HTML12811301Qwen3.5-35B-A3B
Neither model has a result on the other 28 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
Intelligence benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
AA Intelligence Index16.619.3Qwen3.5-35B-A3B
Epoch Capabilities Index144.9142.5DeepSeek V3.2 Exp
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
τ²-Bench Telecom33.9%89.2%Qwen3.5-35B-A3B
TheAgentCompany42.9%Not measuredOnly DeepSeek V3.2 Exp measured
Vending-Bench 2Not measured$463Only Qwen3.5-35B-A3B measured
Neither model has a result on the other 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index32
Reasoning benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
GPQA Diamond (Artificial Analysis run)79.7%84.5%Qwen3.5-35B-A3B
Humanity's Last Exam (Artificial Analysis run)14.9%21.0%Qwen3.5-35B-A3B
GPQA Diamond (Epoch AI run)Not measured83.5%Only Qwen3.5-35B-A3B measured
LMArena Text: hard prompts14291400DeepSeek V3.2 Exp
LMArena Text: expert14221408DeepSeek V3.2 Exp
LMArena Text: hard prompts in English14441411DeepSeek V3.2 Exp
Neither model has a result on the other 15 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
AA Math Index87.7Not measuredOnly DeepSeek V3.2 Exp measured
AIME 2025 (Artificial Analysis run)87.7%Not measuredOnly DeepSeek V3.2 Exp measured
OTIS Mock AIMENot measured70.0%Only Qwen3.5-35B-A3B measured
LMArena Text: math14221404DeepSeek V3.2 Exp
LMArena Text: mathematical work14331410DeepSeek V3.2 Exp
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
Knowledge benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
MMLU Pro (Artificial Analysis run)85.0%Not measuredOnly DeepSeek V3.2 Exp measured
LMArena Text: science14431413DeepSeek V3.2 Exp
LMArena Text: medicine and healthcare14681421DeepSeek V3.2 Exp
LMArena Text: legal and government14191399DeepSeek V3.2 Exp
LMArena Text: business and finance14121394DeepSeek V3.2 Exp
Neither model has a result on the other 12 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
Web design benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
Design Arena: websites1179Not measuredOnly DeepSeek V3.2 Exp measured
Design Arena: UI components1166Not measuredOnly DeepSeek V3.2 Exp measured
Design Arena: game dev1145Not measuredOnly DeepSeek V3.2 Exp measured
Design Arena: data viz1152Not measuredOnly DeepSeek V3.2 Exp measured
Design Arena: 3D1148Not measuredOnly DeepSeek V3.2 Exp measured
Design Arena: all code categories1171Not measuredOnly DeepSeek V3.2 Exp measured
Design Arena: SVG1031Not measuredOnly DeepSeek V3.2 Exp measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index192
General benchmarks for DeepSeek V3.2 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
IFBench54.1%72.5%Qwen3.5-35B-A3B
Long context reasoning (AA-LCR)72.3%72.0%DeepSeek V3.2 Exp
DTBench87.7%80.0%DeepSeek V3.2 Exp
LMCA29.1%29.5%Qwen3.5-35B-A3B
Fiction.LiveBench83.3%Not measuredOnly DeepSeek V3.2 Exp measured
CL-bench13.2%Not measuredOnly DeepSeek V3.2 Exp measured
CL-bench Life9.5%Not measuredOnly DeepSeek V3.2 Exp measured
LMArena Text14251395DeepSeek V3.2 Exp
LMArena Text: instruction following14011379DeepSeek V3.2 Exp
LMArena Text: creative writing14041347DeepSeek V3.2 Exp
LMArena Text: English14411407DeepSeek V3.2 Exp
LMArena Text: non-English14041378DeepSeek V3.2 Exp
LMArena Text: Chinese14611457DeepSeek V3.2 Exp
LMArena Text: German14401367DeepSeek V3.2 Exp
LMArena Text: Japanese13311326DeepSeek V3.2 Exp
LMArena Text: Korean13721355DeepSeek V3.2 Exp
LMArena Text: French14341412DeepSeek V3.2 Exp
LMArena Text: Spanish14411390DeepSeek V3.2 Exp
LMArena Text: Russian14121376DeepSeek V3.2 Exp
LMArena Text: Polish14351385DeepSeek V3.2 Exp
LMArena Text: multi-turn14221390DeepSeek V3.2 Exp
LMArena Text: longer queries14181389DeepSeek V3.2 Exp
LMArena Text: writing and language14071365DeepSeek V3.2 Exp
LMArena Text: entertainment and media14071347DeepSeek 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 Exp and Qwen3.5-35B-A3B
BenchmarkDeepSeek V3.2 ExpQwen3.5-35B-A3BLeader
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 Exp$0.0005Fits in one request
Qwen3.5-35B-A3B$0.0006Fits in one request

DeepSeek V3.2 Exp costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2 Exp$0.015Fits in one request
Qwen3.5-35B-A3B$0.010Fits in one request

Qwen3.5-35B-A3B costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2 Exp$0.058Exceeds the context window
Qwen3.5-35B-A3B$0.038Fits in one request

Qwen3.5-35B-A3B 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 Exp and Qwen3.5-35B-A3B
SpecificationDeepSeek V3.2 ExpQwen3.5-35B-A3B
MakerDeepSeekQwen
WeightsOpen weightsOpen weights
ReleasedSeptember 29, 2025February 25, 2026
Context window164K262K
Input $/Mtok$0.27$0.14
Output $/Mtok$0.41$1.00
Coding index55.07.7
Independent measurements behind it21
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 Exp better than Qwen3.5-35B-A3B for coding?

Pick DeepSeek V3.2 Exp for coding: it scores 55.0 on the Vibe Coding Index against 7.7 for Qwen3.5-35B-A3B. Pick Qwen3.5-35B-A3B when the context window is larger, 262K against 164K tokens, or when what matters is LMArena WebDev: HTML or AA Intelligence Index, where it leads.

Which is cheaper, DeepSeek V3.2 Exp or Qwen3.5-35B-A3B?

DeepSeek V3.2 Exp costs $0.27 per million input tokens and $0.41 per million output tokens; Qwen3.5-35B-A3B costs $0.14 and $1.00. Cost per solved task is known for neither of them.

Which has the bigger context window, DeepSeek V3.2 Exp or Qwen3.5-35B-A3B?

Qwen3.5-35B-A3B, with 262K tokens against 164K.

How many benchmarks is the DeepSeek V3.2 Exp vs Qwen3.5-35B-A3B comparison based on?

40 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.2 Exp's coding index rests on 2 independent benchmarks and Qwen3.5-35B-A3B's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next