Enter opens the highlighted result. Esc closes.

DeepSeek V3.1 Terminus vs Gemma 4 31B

DeepSeek V3.1 Terminus leads Gemma 4 31B on the Artificial Analysis Intelligence Index, 14.8 to 14.7, and Gemma 4 31B costs 2.5x less per output token ($0.40 vs $1.0142 per million). Updated October 21, 2026; 34 benchmarks have a result for both.

Replace DeepSeek V3.1 Terminus Replace Gemma 4 31B
DeepSeek V3.1 TerminusDeepSeek
Coding
No data
Overall
No data
Output $/M
$1.0142
5
Gemma 4 31BGoogle
Coding
39.0
Overall
42.1
Output $/M
$0.40
29
vs boards won of 34 shared

DeepSeek V3.1 Terminus or Gemma 4 31B: which to pick

Only Gemma 4 31B has a coding score, 39.0 on the Vibe Coding Index; DeepSeek V3.1 Terminus has none, so the two cannot be ranked on coding.

Pick DeepSeek V3.1 Terminus for work measured by AA Coding Index or AA Intelligence Index, where it leads.

Model ADeepSeek V3.1 TerminusDeepSeekNot benchmarkedNo coding benchmarkNo cost per solved taskModel BGemma 4 31BGoogle39.0/1002 independent benchmarksNo cost per solved task
Shared results
34
DeepSeek V3.1 Terminus only
11
Gemma 4 31B only
37
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 V3.1 Terminus not benchmarked, Gemma 4 31B 39.0 from 2 benchmarks.
Widest category
Agentic work, 15.5 points apart on average over 2 shared benchmarks.
Widest single gap
τ²-Bench Telecom: DeepSeek V3.1 Terminus 37.1% against Gemma 4 31B 65.5%, 28.4 points apart on a 0 to 100 scale.
Output price
DeepSeek V3.1 Terminus $1.0142 against Gemma 4 31B $0.40 per million output tokens, 2.5x apart.
Context window
DeepSeek V3.1 Terminus takes 164K tokens, Gemma 4 31B 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 V3.1 TerminusBarsGemma 4 31BTies
Coding15
Intelligence10
Agentic work11
Reasoning05
Math01
Knowledge22
General015
All shared boards529
CodingLike-for-likeSix of these boards feed the Vibe Coding Index15
Coding benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
LiveBench codingNot measured49.0%Only Gemma 4 31B measured
LMArena WebDevNot measured1367Only Gemma 4 31B measured
AA Coding Index43.543.4DeepSeek V3.1 Terminus
LiveCodeBench (Artificial Analysis run)79.8%Not measuredOnly DeepSeek V3.1 Terminus measured
SciCode38.0%45.5%Gemma 4 31B
Terminal-Bench Hard (Artificial Analysis run)31.8%36.4%Gemma 4 31B
WeirdMLNot measured52.3%Only Gemma 4 31B measured
ALE-Bench745.2925.5Gemma 4 31B
LMArena Text: coding14251459Gemma 4 31B
LMArena Text: software and IT14261463Gemma 4 31B
LMArena WebDev: ReactNot measured1359Only Gemma 4 31B measured
LMArena WebDev: HTMLNot measured1380Only Gemma 4 31B measured
SciCodeNot measured8.9%Only Gemma 4 31B measured
Neither model has a result on the other 26 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
AA Intelligence Index14.814.7DeepSeek V3.1 Terminus
Epoch Capabilities IndexNot measured143.0Only Gemma 4 31B measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
Agentic work benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
AA Agentic Index6.84.2DeepSeek V3.1 Terminus
LiveBench researchNot measured44.4%Only Gemma 4 31B measured
LiveBench browserNot measured4.0%Only Gemma 4 31B measured
LiveBench long agenticNot measured13.3%Only Gemma 4 31B measured
τ²-Bench Telecom37.1%65.5%Gemma 4 31B
Neither model has a result on the other 36 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
LiveBench reasoningNot measured68.3%Only Gemma 4 31B measured
GPQA Diamond (Artificial Analysis run)79.2%85.7%Gemma 4 31B
Humanity's Last Exam (Artificial Analysis run)16.4%23.6%Gemma 4 31B
GPQA Diamond (Epoch AI run)Not measured75.8%Only Gemma 4 31B measured
LMArena Text: hard prompts14261448Gemma 4 31B
LMArena Text: expertNot measured1464Only Gemma 4 31B measured
LMArena Text: hard prompts in English14321456Gemma 4 31B
LMArena Vision: diagramsNot measured1283Only Gemma 4 31B measured
LMArena Vision: homeworkNot measured1310Only Gemma 4 31B measured
Kagi LLM Benchmark57.4%63.5%Gemma 4 31B
Neither model has a result on the other 12 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index01
Math benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
AA Math Index89.7Not measuredOnly DeepSeek V3.1 Terminus measured
AIME 2025 (Artificial Analysis run)89.7%Not measuredOnly DeepSeek V3.1 Terminus measured
OTIS Mock AIMENot measured73.3%Only Gemma 4 31B measured
LMArena Text: math14021463Gemma 4 31B
LMArena Text: mathematical workNot measured1465Only Gemma 4 31B measured
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index22
Knowledge benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
CaseLaw (Vals AI)Not measured52.6%Only Gemma 4 31B measured
MortgageTax (Vals AI)Not measured61.4%Only Gemma 4 31B measured
MMLU Pro (Artificial Analysis run)85.1%Not measuredOnly DeepSeek V3.1 Terminus measured
SimpleQA VerifiedNot measured10.4%Only Gemma 4 31B measured
LMArena Text: science14511457Gemma 4 31B
LMArena Text: medicine and healthcare14681451DeepSeek V3.1 Terminus
LMArena Text: legal and government14571438DeepSeek V3.1 Terminus
LMArena Text: business and finance14141430Gemma 4 31B
FACTS Benchmark SuiteNot measured45.3%Only Gemma 4 31B measured
Neither model has a result on the other 9 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
Design Arena: websites1187Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: UI components1182Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: game dev1131Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: data viz1167Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: 3D1141Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: all code categories1175Not measuredOnly DeepSeek V3.1 Terminus measured
Design Arena: SVG1065Not measuredOnly DeepSeek V3.1 Terminus measured
LiveBench designNot measured58.4%Only Gemma 4 31B measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index015
General benchmarks for DeepSeek V3.1 Terminus and Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
LiveBench data analysisNot measured53.3%Only Gemma 4 31B measured
SAGE (Vals AI)Not measured55.0%Only Gemma 4 31B measured
IFBench57.0%75.6%Gemma 4 31B
Long context reasoning (AA-LCR)69.3%69.7%Gemma 4 31B
DTBench81.3%82.7%Gemma 4 31B
LMCA28.6%39.3%Gemma 4 31B
Surface Evolver BenchNot measured30.6%Only Gemma 4 31B measured
LMArena Text14191443Gemma 4 31B
LMArena Text: instruction following14041433Gemma 4 31B
LMArena Text: creative writing14031417Gemma 4 31B
LMArena VisionNot measured1276Only Gemma 4 31B measured
LMArena Text: English14291451Gemma 4 31B
LMArena Text: non-English14081430Gemma 4 31B
LMArena Text: ChineseNot measured1475Only Gemma 4 31B measured
LMArena Text: FrenchNot measured1436Only Gemma 4 31B measured
LMArena Text: SpanishNot measured1442Only Gemma 4 31B measured
LMArena Text: Russian14351460Gemma 4 31B
LMArena Text: Polish14271433Gemma 4 31B
LMArena Text: multi-turn14111452Gemma 4 31B
LMArena Text: longer queries14221446Gemma 4 31B
LMArena Text: writing and language14091432Gemma 4 31B
LMArena Text: entertainment and media13911405Gemma 4 31B
LMArena Vision: EnglishNot measured1271Only Gemma 4 31B measured
LMArena Vision: ChineseNot measured1326Only Gemma 4 31B measured
LMArena Vision: reading textNot measured1282Only Gemma 4 31B measured
LMArena Vision: humorNot measured1280Only Gemma 4 31B measured
LMArena Vision: captioningNot measured1218Only Gemma 4 31B measured
LMArena Vision: recognizing thingsNot measured1226Only Gemma 4 31B measured
LMArena Vision: creative writingNot measured1285Only Gemma 4 31B measured
LMArena DocumentNot measured1425Only Gemma 4 31B measured
Neither model has a result on the other 16 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 Gemma 4 31B
BenchmarkDeepSeek V3.1 TerminusGemma 4 31BLeader
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
Gemma 4 31B$0.0003Fits in one request

Gemma 4 31B costs less

Repository review

50K input + 3K output tokens

DeepSeek V3.1 Terminus$0.018Fits in one request
Gemma 4 31B$0.0082Fits in one request

Gemma 4 31B costs less

Agent loop

200K input + 10K output tokens

DeepSeek V3.1 Terminus$0.071Exceeds the context window
Gemma 4 31B$0.032Fits in one request

Gemma 4 31B 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 Gemma 4 31B
SpecificationDeepSeek V3.1 TerminusGoogle Gemma 4 31B
MakerDeepSeekGoogle
WeightsOpen weightsOpen weights
ReleasedSeptember 22, 2025April 2, 2026
Context window164K262K
Input $/Mtok$0.3063$0.14
Output $/Mtok$1.0142$0.40
Coding indexNot benchmarked39.0
Independent measurements behind it02
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V3.1 Terminus better than Gemma 4 31B for coding?

Only Gemma 4 31B has a coding score, 39.0 on the Vibe Coding Index; DeepSeek V3.1 Terminus has none, so the two cannot be ranked on coding. Pick DeepSeek V3.1 Terminus for work measured by AA Coding Index or AA Intelligence Index, where it leads.

Which is cheaper, DeepSeek V3.1 Terminus or Gemma 4 31B?

DeepSeek V3.1 Terminus costs $0.3063 per million input tokens and $1.0142 per million output tokens; Gemma 4 31B costs $0.14 and $0.40. Cost per solved task is known for neither of them.

Which has the bigger context window, DeepSeek V3.1 Terminus or Gemma 4 31B?

Gemma 4 31B, with 262K tokens against 164K.

How many benchmarks is the DeepSeek V3.1 Terminus vs Gemma 4 31B comparison based on?

34 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.1 Terminus's coding index rests on 0 independent benchmarks and Gemma 4 31B's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next