Enter opens the highlighted result. Esc closes.

DeepSeek V3.2 vs gpt-oss-20b

DeepSeek V3.2 leads gpt-oss-20b on the Artificial Analysis Intelligence Index, 21.5 to 10.0, and gpt-oss-20b costs 5.9x less per output token ($0.15 vs $0.88 per million). Updated October 21, 2026; 54 benchmarks have a result for both.

Replace DeepSeek V3.2 Replace gpt-oss-20b
DeepSeek V3.2DeepSeek
Coding
30.0
Overall
21.5
Output $/M
$0.88
50
gpt-oss-20bOpenAI
Coding
No data
Overall
10.0
Output $/M
$0.15
4
vs boards won of 54 shared

DeepSeek V3.2 or gpt-oss-20b: which to pick

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; gpt-oss-20b has none, so the two cannot be ranked on coding.

Pick gpt-oss-20b for output tokens 5.9x cheaper ($0.15 vs $0.88 per million) and work measured by AIME or MGSM (Vals AI), where it leads.

Model ADeepSeek V3.2DeepSeek30.0/1003 independent benchmarks$0.99 per solved taskModel Bgpt-oss-20bOpenAINot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
54
DeepSeek V3.2 only
21
gpt-oss-20b 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, gpt-oss-20b not benchmarked.
Widest category
Web design, 34.3 points apart on average over 2 shared benchmarks.
Widest single gap
Design Arena: websites: DeepSeek V3.2 1175 against gpt-oss-20b 852, 40.4 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.2 $0.88 against gpt-oss-20b $0.15 per million output tokens, 5.9x apart.
Context window
DeepSeek V3.2 takes 164K tokens, gpt-oss-20b 131K.

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 V3.2Barsgpt-oss-20bTies
Coding60
Intelligence20
Agentic work10
Reasoning60
Math42
Knowledge101
Web design20
General191
All shared boards504
CodingLike-for-likeSix of these boards feed the Vibe Coding Index60
Coding benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
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%80.4%DeepSeek V3.2
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%77.7%DeepSeek V3.2
SciCodeNot measured38.9%Only gpt-oss-20b measured
Terminal-Bench Hard (Artificial Analysis run)35.6%10.6%DeepSeek V3.2
Terminal-Bench39.6%3.4%DeepSeek V3.2
WeirdMLNot measured40.9%Only gpt-oss-20b measured
ALE-BenchNot measured566.0Only gpt-oss-20b measured
LMArena Text: coding14531307DeepSeek V3.2
LMArena Text: software and IT14491311DeepSeek 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 21 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
AA Intelligence Index21.510.0DeepSeek V3.2
Epoch Capabilities Index146.1137.7DeepSeek V3.2
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
τ²-Bench Telecom90.6%60.2%DeepSeek V3.2
APEX-Agents21.3%Not measuredOnly DeepSeek V3.2 measured
Vending-Bench 2$1,034Not measuredOnly DeepSeek V3.2 measured
Neither model has a result on the other 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
GPQA Diamond80.3%68.9%DeepSeek V3.2
GPQA Diamond (Artificial Analysis run)84.0%68.8%DeepSeek V3.2
Humanity's Last Exam (Artificial Analysis run)24.6%11.0%DeepSeek V3.2
GPQA Diamond (Epoch AI run)Not measured60.8%Only gpt-oss-20b measured
ARC-AGI57.0%Not measuredOnly DeepSeek V3.2 measured
ARC-AGI-24.0%Not measuredOnly DeepSeek V3.2 measured
LMArena Text: hard prompts14341274DeepSeek V3.2
LMArena Text: expert14361258DeepSeek V3.2
LMArena Text: hard prompts in English14461292DeepSeek V3.2
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index42
Math benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
AIME84.6%86.0%gpt-oss-20b
MATH 500 (Vals AI)Not measured94.2%Only gpt-oss-20b measured
MGSM (Vals AI)86.0%89.0%gpt-oss-20b
AA Math Index92.089.3DeepSeek V3.2
AIME 2025 (Artificial Analysis run)92.0%89.3%DeepSeek V3.2
OTIS Mock AIMENot measured65.3%Only gpt-oss-20b measured
LMArena Text: math14351317DeepSeek V3.2
LMArena Text: mathematical work14401305DeepSeek V3.2
Neither model has a result on the other 5 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index101
Knowledge benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
MMLU Pro84.9%71.6%DeepSeek V3.2
LegalBench (Vals AI)76.1%70.8%DeepSeek V3.2
CaseLaw (Vals AI)55.4%43.8%DeepSeek V3.2
CorpFin (Vals AI)51.0%53.1%gpt-oss-20b
TaxEval (Vals AI)68.2%63.7%DeepSeek V3.2
MedQA (Vals AI)93.9%82.9%DeepSeek V3.2
MMLU Pro (Artificial Analysis run)86.2%74.8%DeepSeek V3.2
LMArena Text: science14461296DeepSeek V3.2
LMArena Text: medicine and healthcare14401310DeepSeek V3.2
LMArena Text: legal and government14251280DeepSeek V3.2
LMArena Text: business and finance14101284DeepSeek V3.2
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index20
Web design benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
Design Arena: websites1175852DeepSeek V3.2
Design Arena: UI components1149Not measuredOnly DeepSeek V3.2 measured
Design Arena: game dev1132Not measuredOnly DeepSeek V3.2 measured
Design Arena: data viz1158933DeepSeek V3.2
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 index191
General benchmarks for DeepSeek V3.2 and gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
IFBench60.7%65.1%gpt-oss-20b
Long context reasoning (AA-LCR)73.3%34.7%DeepSeek V3.2
DTBench85.6%68.0%DeepSeek V3.2
LMCA28.8%14.5%DeepSeek V3.2
CL-bench12.4%Not measuredOnly DeepSeek V3.2 measured
CL-bench Life7.4%Not measuredOnly DeepSeek V3.2 measured
LMArena Text14251288DeepSeek V3.2
LMArena Text: instruction following14131236DeepSeek V3.2
LMArena Text: creative writing14001201DeepSeek V3.2
LMArena Text: English14391303DeepSeek V3.2
LMArena Text: non-English14091269DeepSeek V3.2
LMArena Text: Chinese14551315DeepSeek V3.2
LMArena Text: German14131255DeepSeek V3.2
LMArena Text: Japanese13741244DeepSeek V3.2
LMArena Text: Korean13701236DeepSeek V3.2
LMArena Text: French1429Not measuredOnly DeepSeek V3.2 measured
LMArena Text: Spanish14241267DeepSeek V3.2
LMArena Text: Russian14241278DeepSeek V3.2
LMArena Text: Polish14191252DeepSeek V3.2
LMArena Text: multi-turn14271268DeepSeek V3.2
LMArena Text: longer queries14281250DeepSeek V3.2
LMArena Text: writing and language14071238DeepSeek V3.2
LMArena Text: entertainment and media13951216DeepSeek V3.2
Neither model has a result on the other 23 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 gpt-oss-20b
BenchmarkDeepSeek V3.2gpt-oss-20bLeader
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
gpt-oss-20b$0.0001Fits in one request

gpt-oss-20b costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.2$0.017Fits in one request
gpt-oss-20b$0.0019Fits in one request

gpt-oss-20b costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.2$0.067Exceeds the context window
gpt-oss-20b$0.0075Exceeds the context window

gpt-oss-20b 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 gpt-oss-20b
SpecificationDeepSeek V3.2OpenAI gpt-oss-20b
MakerDeepSeekOpenAI
WeightsOpen weightsOpen weights
ReleasedDecember 1, 2025August 5, 2025
Context window164K131K
Input $/Mtok$0.29$0.03
Output $/Mtok$0.88$0.15
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 gpt-oss-20b for coding?

Only DeepSeek V3.2 has a coding score, 30.0 on the Vibe Coding Index; gpt-oss-20b has none, so the two cannot be ranked on coding. Pick gpt-oss-20b for output tokens 5.9x cheaper ($0.15 vs $0.88 per million) and work measured by AIME or MGSM (Vals AI), where it leads.

Which is cheaper, DeepSeek V3.2 or gpt-oss-20b?

DeepSeek V3.2 costs $0.29 per million input tokens and $0.88 per million output tokens; gpt-oss-20b costs $0.03 and $0.15. Cost per solved task is known for DeepSeek V3.2 only.

Which has the bigger context window, DeepSeek V3.2 or gpt-oss-20b?

DeepSeek V3.2, with 164K tokens against 131K.

How many benchmarks is the DeepSeek V3.2 vs gpt-oss-20b comparison based on?

54 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.2's coding index rests on 3 independent benchmarks and gpt-oss-20b's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next