Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs Qwen2.5 72B Instruct

DeepSeek V3.2 leads Qwen2.5 72B Instruct on the Artificial Analysis Intelligence Index, 21.5 to 7.7, and Qwen2.5 72B Instruct costs 2.2x less per output token ($0.40 vs $0.88 per million). Updated October 21, 2026; 40 benchmarks have a result for both.

Replace DeepSeek V3.2 Replace Qwen2.5 72B Instruct
DeepSeek V3.2DeepSeek
Coding
30.0
Overall
37.9
Output $/M
$0.88
40
Qwen2.5 72B InstructQwen
Coding
No data
Overall
26.5
Output $/M
$0.40
0
vs boards won of 40 shared

DeepSeek V3.2 or Qwen2.5 72B Instruct: which to pick

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; Qwen2.5 72B Instruct has none, so the two cannot be ranked on coding.

Pick Qwen2.5 72B Instruct for output tokens 2.2x cheaper ($0.40 vs $0.88 per million).

Model ADeepSeek V3.2DeepSeek30.0/1003 independent benchmarks$0.99 per solved taskModel BQwen2.5 72B InstructQwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
40
DeepSeek V3.2 only
35
Qwen2.5 72B Instruct only
13
Like-for-like categories
7 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, Qwen2.5 72B Instruct not benchmarked.
Widest category
Agentic work, 56.1 points apart on average over 1 shared benchmark.
Widest single gap
AA Math Index: DeepSeek V3.2 92.0 against Qwen2.5 72B Instruct 14.0, 78.0 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.88 against Qwen2.5 72B Instruct $0.40 per million output tokens, 2.2x apart.
Context window
DeepSeek V3.2 takes 164K tokens, Qwen2.5 72B Instruct 33K.

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 V3.2BarsQwen2.5 72B InstructTies
Coding40
Intelligence20
Agentic work10
Reasoning50
Math40
Knowledge50
General190
All shared boards400
CodingLike-for-likeSix of these boards feed the Vibe Coding Index40
Coding benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
SWE-bench Verified70.0%Not measuredOnly DeepSeek V3.2 measured
SWE-bench Multilingual59.0%Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev1362Not measuredOnly DeepSeek V3.2 measured
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%27.6%DeepSeek V3.2
Terminal-Bench Hard (Artificial Analysis run)35.6%4.5%DeepSeek V3.2
Terminal-Bench39.6%Not measuredOnly DeepSeek V3.2 measured
WeirdMLNot measured16.0%Only Qwen2.5 72B Instruct measured
LMArena Text: coding14531292DeepSeek V3.2
LMArena Text: software and IT14491292DeepSeek V3.2
LMArena WebDev: React1350Not measuredOnly DeepSeek V3.2 measured
LMArena WebDev: HTML1371Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
AA Intelligence Index21.57.7DeepSeek V3.2
Epoch Capabilities Index146.1129.0DeepSeek V3.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
τ²-Bench Telecom90.6%34.5%DeepSeek V3.2
APEX-Agents21.3%Not measuredOnly DeepSeek V3.2 measured
METR time horizonsNot measured35.8%Only Qwen2.5 72B Instruct measured
BALROGNot measured16.2%Only Qwen2.5 72B Instruct measured
TheAgentCompanyNot measured5.7%Only Qwen2.5 72B Instruct measured
Vending-Bench 2$1,034Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 35 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index50
Reasoning benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
GPQA Diamond80.3%Not measuredOnly DeepSeek V3.2 measured
GPQA Diamond (Artificial Analysis run)84.0%49.1%DeepSeek V3.2
Humanity's Last Exam (Artificial Analysis run)24.6%3.6%DeepSeek V3.2
GPQA Diamond (Epoch AI run)Not measured49.1%Only Qwen2.5 72B Instruct measured
ARC-AGI57.0%Not measuredOnly DeepSeek V3.2 measured
ARC-AGI-24.0%Not measuredOnly DeepSeek V3.2 measured
BIG-Bench HardNot measured79.8%Only Qwen2.5 72B Instruct measured
LMArena Text: hard prompts14341271DeepSeek V3.2
LMArena Text: expert14361245DeepSeek V3.2
LMArena Text: hard prompts in English14461279DeepSeek V3.2
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index40
Math benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
AIME84.6%Not measuredOnly DeepSeek V3.2 measured
MGSM (Vals AI)86.0%Not measuredOnly DeepSeek V3.2 measured
AA Math Index92.014.0DeepSeek V3.2
AIME 2025 (Artificial Analysis run)92.0%14.0%DeepSeek V3.2
OTIS Mock AIMENot measured8.1%Only Qwen2.5 72B Instruct measured
MATH level 5Not measured63.2%Only Qwen2.5 72B Instruct measured
LMArena Text: math14351283DeepSeek V3.2
LMArena Text: mathematical work14401290DeepSeek V3.2
Neither model has a result on the other 5 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
Knowledge benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
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%72.0%DeepSeek V3.2
MMLUNot measured85.3%Only Qwen2.5 72B Instruct measured
AI2 Reasoning ChallengeNot measured94.5%Only Qwen2.5 72B Instruct measured
LMArena Text: science14461275DeepSeek V3.2
LMArena Text: medicine and healthcare14401258DeepSeek V3.2
LMArena Text: legal and government14251288DeepSeek V3.2
LMArena Text: business and finance14101253DeepSeek V3.2
Neither model has a result on the other 5 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
Design Arena: websites1175Not measuredOnly DeepSeek V3.2 measured
Design Arena: UI components1149Not measuredOnly DeepSeek V3.2 measured
Design Arena: game dev1132Not measuredOnly DeepSeek V3.2 measured
Design Arena: data viz1158Not measuredOnly DeepSeek V3.2 measured
Design Arena: 3D1133Not measuredOnly DeepSeek V3.2 measured
Design Arena: all code categories1164Not measuredOnly DeepSeek V3.2 measured
Design Arena: SVG1032Not measuredOnly DeepSeek V3.2 measured
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 index190
General benchmarks for DeepSeek V3.2 and Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
IFBench60.7%36.9%DeepSeek V3.2
Long context reasoning (AA-LCR)73.3%Not measuredOnly DeepSeek V3.2 measured
DTBench85.6%62.9%DeepSeek V3.2
LMCA28.8%13.4%DeepSeek V3.2
CL-bench12.4%Not measuredOnly DeepSeek V3.2 measured
HellaSwagNot measured84.8%Only Qwen2.5 72B Instruct measured
WinoGrandeNot measured82.3%Only Qwen2.5 72B Instruct measured
PIQANot measured82.6%Only Qwen2.5 72B Instruct measured
CL-bench Life7.4%Not measuredOnly DeepSeek V3.2 measured
LMArena Text14251269DeepSeek V3.2
LMArena Text: instruction following14131254DeepSeek V3.2
LMArena Text: creative writing14001221DeepSeek V3.2
LMArena Text: English14391283DeepSeek V3.2
LMArena Text: non-English14091252DeepSeek V3.2
LMArena Text: Chinese14551273DeepSeek V3.2
LMArena Text: German14131234DeepSeek V3.2
LMArena Text: Japanese13741180DeepSeek V3.2
LMArena Text: Korean13701188DeepSeek V3.2
LMArena Text: French14291280DeepSeek V3.2
LMArena Text: Spanish14241255DeepSeek V3.2
LMArena Text: Russian14241264DeepSeek V3.2
LMArena Text: Polish1419Not measuredOnly DeepSeek V3.2 measured
LMArena Text: multi-turn14271272DeepSeek V3.2
LMArena Text: longer queries14281281DeepSeek V3.2
LMArena Text: writing and language14071247DeepSeek V3.2
LMArena Text: entertainment and media13951211DeepSeek V3.2
Neither model has a result on the other 20 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 Qwen2.5 72B Instruct
BenchmarkDeepSeek V3.2Qwen2.5 72B InstructLeader
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
Qwen2.5 72B Instruct$0.0006Fits in one request

Qwen2.5 72B Instruct costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2$0.017Fits in one request
Qwen2.5 72B Instruct$0.019Exceeds the context window

DeepSeek V3.2 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2$0.067Exceeds the context window
Qwen2.5 72B Instruct$0.076Exceeds the context window

DeepSeek V3.2 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 Qwen2.5 72B Instruct
SpecificationDeepSeek V3.2Qwen2.5 72B Instruct
MakerDeepSeekQwen
WeightsOpen weightsOpen weights
ReleasedDecember 1, 2025September 19, 2024
Context window164K33K
Input $/Mtok$0.29$0.36
Output $/Mtok$0.88$0.40
Coding index30.0Not benchmarked
Independent measurements behind it30
Cost per solved task$0.99Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.2 better than Qwen2.5 72B Instruct for coding?

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; Qwen2.5 72B Instruct has none, so the two cannot be ranked on coding. Pick Qwen2.5 72B Instruct for output tokens 2.2x cheaper ($0.40 vs $0.88 per million).

Which is cheaper, DeepSeek V3.2 or Qwen2.5 72B Instruct?

DeepSeek V3.2 costs $0.29 per million input tokens and $0.88 per million output tokens; Qwen2.5 72B Instruct costs $0.36 and $0.40. Cost per solved task is known for DeepSeek V3.2 only.

Which has the bigger context window, DeepSeek V3.2 or Qwen2.5 72B Instruct?

DeepSeek V3.2, with 164K tokens against 33K.

How many benchmarks is the DeepSeek V3.2 vs Qwen2.5 72B Instruct comparison based on?

40 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.2's coding index rests on 3 independent benchmarks and Qwen2.5 72B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next