Enter opens the highlighted result. Esc closes.

DeepSeek V3.1 Terminus vs Qwen3 30B A3B

DeepSeek V3.1 Terminus leads Qwen3 30B A3B on the Artificial Analysis Intelligence Index, 14.8 to 7.6, and Qwen3 30B A3B costs 2.0x less per output token ($0.50 vs $1.0142 per million). Updated October 22, 2026; 38 benchmarks have a result for both.

Replace DeepSeek V3.1 Terminus Replace Qwen3 30B A3B
DeepSeek V3.1 TerminusDeepSeek
Coding
No data
Overall
No data
Output $/M
$1.0142
38
Qwen3 30B A3BQwen
Coding
No data
Overall
38.2
Output $/M
$0.50
0
vs boards won of 38 shared

DeepSeek V3.1 Terminus or Qwen3 30B A3B: which to pick

Neither model has a coding score yet, so the two cannot be ranked on coding.

Pick DeepSeek V3.1 Terminus for a larger context window (164K against 41K tokens).

Model ADeepSeek V3.1 TerminusDeepSeekNot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen3 30B A3BQwenNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
38
DeepSeek V3.1 Terminus only
7
Qwen3 30B A3B only
14
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 V3.1 Terminus not benchmarked, Qwen3 30B A3B not benchmarked.
Widest category
Web design, 28.5 points apart on average over 5 shared benchmarks.
Widest single gap
Long context reasoning (AA-LCR): DeepSeek V3.1 Terminus 69.3% against Qwen3 30B A3B 0.0%, 69.3 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.1 Terminus $1.0142 against Qwen3 30B A3B $0.50 per million output tokens, 2.0x apart.
Context window
DeepSeek V3.1 Terminus takes 164K tokens, Qwen3 30B A3B 41K.

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 V3.1 TerminusBarsQwen3 30B A3BTies
Coding40
Intelligence10
Agentic work10
Reasoning40
Math30
Knowledge50
Web design50
General150
All shared boards380
CodingLike-for-likeSix of these boards feed the Vibe Coding Index40
Coding benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
AA Coding Index43.5Not measuredOnly DeepSeek V3.1 Terminus measured
LiveCodeBench (Artificial Analysis run)79.8%50.6%DeepSeek V3.1 Terminus
SciCode38.0%Not measuredOnly DeepSeek V3.1 Terminus measured
Terminal-Bench Hard (Artificial Analysis run)31.8%6.8%DeepSeek V3.1 Terminus
WeirdMLNot measured29.8%Only Qwen3 30B A3B measured
ALE-Bench745.2Not measuredOnly DeepSeek V3.1 Terminus measured
LMArena Text: coding14251337DeepSeek V3.1 Terminus
LMArena Text: software and IT14261340DeepSeek V3.1 Terminus
Neither model has a result on the other 31 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
AA Intelligence Index14.87.6DeepSeek V3.1 Terminus
Epoch Capabilities IndexNot measured137.3Only Qwen3 30B A3B measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
AA Agentic Index6.8Not measuredOnly DeepSeek V3.1 Terminus measured
τ²-Bench Telecom37.1%26.0%DeepSeek V3.1 Terminus
Neither model has a result on the other 39 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index40
Reasoning benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
GPQA Diamond (Artificial Analysis run)79.2%61.6%DeepSeek V3.1 Terminus
Humanity's Last Exam (Artificial Analysis run)16.4%6.2%DeepSeek V3.1 Terminus
GPQA Diamond (Epoch AI run)Not measured61.7%Only Qwen3 30B A3B measured
LMArena Text: hard prompts14261314DeepSeek V3.1 Terminus
LMArena Text: expertNot measured1314Only Qwen3 30B A3B measured
LMArena Text: hard prompts in English14321329DeepSeek V3.1 Terminus
Kagi LLM Benchmark57.4%Not measuredOnly DeepSeek V3.1 Terminus measured
Neither model has a result on the other 15 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index30
Math benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
AA Math Index89.772.3DeepSeek V3.1 Terminus
AIME 2025 (Artificial Analysis run)89.7%72.3%DeepSeek V3.1 Terminus
OTIS Mock AIMENot measured62.8%Only Qwen3 30B A3B measured
LMArena Text: math14021355DeepSeek V3.1 Terminus
LMArena Text: mathematical workNot measured1363Only Qwen3 30B A3B measured
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
Knowledge benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
MMLU Pro (Artificial Analysis run)85.1%77.7%DeepSeek V3.1 Terminus
LMArena Text: science14511321DeepSeek V3.1 Terminus
LMArena Text: medicine and healthcare14681318DeepSeek V3.1 Terminus
LMArena Text: legal and government14571314DeepSeek V3.1 Terminus
LMArena Text: business and finance14141309DeepSeek V3.1 Terminus
Neither model has a result on the other 13 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index50
Web design benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
Design Arena: websites1187954DeepSeek V3.1 Terminus
Design Arena: UI components1182943DeepSeek V3.1 Terminus
Design Arena: game dev1131898DeepSeek V3.1 Terminus
Design Arena: data viz1168967DeepSeek V3.1 Terminus
Design Arena: 3D1141Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: all code categories1175941DeepSeek V3.1 Terminus
Design Arena: SVG1065Not measuredOnly DeepSeek V3.1 Terminus measured
Neither model has a result on the other 2 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index150
General benchmarks for DeepSeek V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B A3BLeader
IFBench57.0%41.5%DeepSeek V3.1 Terminus
Long context reasoning (AA-LCR)69.3%0.0%DeepSeek V3.1 Terminus
DTBench81.3%60.3%DeepSeek V3.1 Terminus
LMCA28.6%15.8%DeepSeek V3.1 Terminus
Fiction.LiveBenchNot measured40.6%Only Qwen3 30B A3B measured
Creative writing (Lech Mazur)Not measured75.3%Only Qwen3 30B A3B measured
LMArena Text14191317DeepSeek V3.1 Terminus
LMArena Text: instruction following14041283DeepSeek V3.1 Terminus
LMArena Text: creative writing14031271DeepSeek V3.1 Terminus
LMArena Text: English14291335DeepSeek V3.1 Terminus
LMArena Text: non-English14081295DeepSeek V3.1 Terminus
LMArena Text: ChineseNot measured1347Only Qwen3 30B A3B measured
LMArena Text: GermanNot measured1307Only Qwen3 30B A3B measured
LMArena Text: JapaneseNot measured1254Only Qwen3 30B A3B measured
LMArena Text: KoreanNot measured1261Only Qwen3 30B A3B measured
LMArena Text: FrenchNot measured1352Only Qwen3 30B A3B measured
LMArena Text: SpanishNot measured1317Only Qwen3 30B A3B measured
LMArena Text: Russian14351291DeepSeek V3.1 Terminus
LMArena Text: Polish14271299DeepSeek V3.1 Terminus
LMArena Text: multi-turn14111307DeepSeek V3.1 Terminus
LMArena Text: longer queries14221312DeepSeek V3.1 Terminus
LMArena Text: writing and language14091290DeepSeek V3.1 Terminus
LMArena Text: entertainment and media13911258DeepSeek V3.1 Terminus
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 V3.1 Terminus and Qwen3 30B A3B
BenchmarkDeepSeek V3.1 TerminusQwen3 30B 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 V3.1 Terminus$0.0008Fits in one request
Qwen3 30B A3B$0.0004Fits in one request

Qwen3 30B A3B costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.1 Terminus$0.018Fits in one request
Qwen3 30B A3B$0.0075Exceeds the context window

Qwen3 30B A3B costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.1 Terminus$0.071Exceeds the context window
Qwen3 30B A3B$0.029Exceeds the context window

Qwen3 30B 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 V3.1 Terminus and Qwen3 30B A3B
SpecificationDeepSeek V3.1 TerminusQwen3 30B A3B
MakerDeepSeekQwen
WeightsOpen weightsOpen weights
ReleasedSeptember 22, 2025April 28, 2025
Context window164K41K
Input $/Mtok$0.3063$0.12
Output $/Mtok$1.0142$0.50
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.1 Terminus better than Qwen3 30B A3B for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick DeepSeek V3.1 Terminus for a larger context window (164K against 41K tokens).

Which is cheaper, DeepSeek V3.1 Terminus or Qwen3 30B A3B?

DeepSeek V3.1 Terminus costs $0.3063 per million input tokens and $1.0142 per million output tokens; Qwen3 30B A3B costs $0.12 and $0.50. Cost per solved task is known for neither of them.

Which has the bigger context window, DeepSeek V3.1 Terminus or Qwen3 30B A3B?

DeepSeek V3.1 Terminus, with 164K tokens against 41K.

How many benchmarks is the DeepSeek V3.1 Terminus vs Qwen3 30B A3B comparison based on?

38 benchmarks have a result for both models as of October 22, 2026. DeepSeek V3.1 Terminus's coding index rests on 0 independent benchmarks and Qwen3 30B A3B's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next