Enter opens the highlighted result. Esc closes.

Gemma 3 12B vs Nemotron 3 Nano 30B A3B

Nemotron 3 Nano 30B A3B leads Gemma 3 12B on the Artificial Analysis Intelligence Index, 8.9 to 3.8, and Gemma 3 12B costs 1.4x less per output token ($0.1625 vs $0.22 per million). Updated October 22, 2026; 36 benchmarks have a result for both.

Replace Gemma 3 12B Replace Nemotron 3 Nano 30B A3B
Gemma 3 12BGoogle
Coding
No data
Overall
28.0
Output $/M
$0.1625
10
Nemotron 3 Nano 30B A3BNVIDIA
Coding
No data
Overall
No data
Output $/M
$0.22
26
vs boards won of 36 shared

Gemma 3 12B or Nemotron 3 Nano 30B A3B: which to pick

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

Pick Gemma 3 12B for output tokens 1.4x cheaper ($0.1625 vs $0.22 per million) and work measured by AA Coding Index or LiveCodeBench (Artificial Analysis run), where it leads.

Model AGemma 3 12BGoogleNot benchmarkedNo coding benchmarkNo cost per solved taskModel BNemotron 3 Nano 30B A3BNVIDIANot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
36
Gemma 3 12B only
7
Nemotron 3 Nano 30B A3B only
6
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
Gemma 3 12B not benchmarked, Nemotron 3 Nano 30B A3B not benchmarked.
Widest category
Math, 40.6 points apart on average over 4 shared benchmarks.
Widest single gap
AA Math Index: Gemma 3 12B 18.3 against Nemotron 3 Nano 30B A3B 91.0, 72.7 points apart on a 0 to 100 scale.
Output price
Gemma 3 12B $0.1625 against Nemotron 3 Nano 30B A3B $0.22 per million output tokens, 1.4x apart.
Context window
Gemma 3 12B takes 131K tokens, Nemotron 3 Nano 30B 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
CategoryGemma 3 12BBarsNemotron 3 Nano 30B A3BTies
Coding06
Intelligence01
Agentic work02
Reasoning05
Math04
Knowledge32
General76
All shared boards1026
CodingLike-for-likeSix of these boards feed the Vibe Coding Index06
Coding benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
AA Coding Index5.814.4Nemotron 3 Nano 30B A3B
LiveCodeBench (Artificial Analysis run)13.7%74.1%Nemotron 3 Nano 30B A3B
SciCode16.4%30.6%Nemotron 3 Nano 30B A3B
Terminal-Bench Hard (Artificial Analysis run)0.8%13.6%Nemotron 3 Nano 30B A3B
LMArena Text: coding12811378Nemotron 3 Nano 30B A3B
LMArena Text: software and IT13131383Nemotron 3 Nano 30B A3B
SciCode0.0%Not measuredOnly Gemma 3 12B measured
Neither model has a result on the other 36 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
AA Intelligence Index3.88.9Nemotron 3 Nano 30B A3B
Epoch Capabilities Index123.5Not measuredOnly Gemma 3 12B measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
Agentic work benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
AA Agentic Index0.11.0Nemotron 3 Nano 30B A3B
τ²-Bench Telecom10.8%40.9%Nemotron 3 Nano 30B A3B
Berkeley Function Calling Leaderboard (BFCL)30.4%Not measuredOnly Gemma 3 12B measured
Neither model has a result on the other 39 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
GPQA Diamond (Artificial Analysis run)34.9%75.7%Nemotron 3 Nano 30B A3B
Humanity's Last Exam (Artificial Analysis run)4.2%11.4%Nemotron 3 Nano 30B A3B
GPQA Diamond (Epoch AI run)39.5%Not measuredOnly Gemma 3 12B measured
LMArena Text: hard prompts13091348Nemotron 3 Nano 30B A3B
LMArena Text: expert12481355Nemotron 3 Nano 30B A3B
LMArena Text: hard prompts in English13181376Nemotron 3 Nano 30B A3B
Neither model has a result on the other 17 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index04
Math benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
AA Math Index18.391.0Nemotron 3 Nano 30B A3B
AIME 2025 (Artificial Analysis run)18.3%91.0%Nemotron 3 Nano 30B A3B
OTIS Mock AIME16.7%Not measuredOnly Gemma 3 12B measured
LMArena Text: math13071371Nemotron 3 Nano 30B A3B
LMArena Text: mathematical work13361375Nemotron 3 Nano 30B A3B
Neither model has a result on the other 9 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index32
Knowledge benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
MMLU Pro (Artificial Analysis run)59.5%79.4%Nemotron 3 Nano 30B A3B
LMArena Text: science13661353Gemma 3 12B
LMArena Text: medicine and healthcare13341349Nemotron 3 Nano 30B A3B
LMArena Text: legal and government13811341Gemma 3 12B
LMArena Text: business and finance13761346Gemma 3 12B
Neither model has a result on the other 15 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index76
General benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 30B A3BLeader
IFBench36.7%71.1%Nemotron 3 Nano 30B A3B
Long context reasoning (AA-LCR)8.3%38.0%Nemotron 3 Nano 30B A3B
DTBench48.8%Not measuredOnly Gemma 3 12B measured
LMCA4.5%Not measuredOnly Gemma 3 12B measured
LMArena Text13341348Nemotron 3 Nano 30B A3B
LMArena Text: instruction following12991303Nemotron 3 Nano 30B A3B
LMArena Text: creative writing13311269Gemma 3 12B
LMArena Text: English13491377Nemotron 3 Nano 30B A3B
LMArena Text: non-English13181320Nemotron 3 Nano 30B A3B
LMArena Text: ChineseNot measured1383Only Nemotron 3 Nano 30B A3B measured
LMArena Text: German13701348Gemma 3 12B
LMArena Text: JapaneseNot measured1268Only Nemotron 3 Nano 30B A3B measured
LMArena Text: KoreanNot measured1269Only Nemotron 3 Nano 30B A3B measured
LMArena Text: FrenchNot measured1396Only Nemotron 3 Nano 30B A3B measured
LMArena Text: SpanishNot measured1376Only Nemotron 3 Nano 30B A3B measured
LMArena Text: Russian13351304Gemma 3 12B
LMArena Text: PolishNot measured1303Only Nemotron 3 Nano 30B A3B measured
LMArena Text: multi-turn13341319Gemma 3 12B
LMArena Text: longer queries13171300Gemma 3 12B
LMArena Text: writing and language13071283Gemma 3 12B
LMArena Text: entertainment and media12971293Gemma 3 12B
Neither model has a result on the other 27 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemma 3 12B and Nemotron 3 Nano 30B A3B
BenchmarkGemma 3 12BNemotron 3 Nano 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

Gemma 3 12B$0.0002Fits in one request
Nemotron 3 Nano 30B A3B$0.0002Fits in one request

Gemma 3 12B costs less

Repository review

50K input + 3K output tokens

Gemma 3 12B$0.0042Fits in one request
Nemotron 3 Nano 30B A3B$0.0034Fits in one request

Nemotron 3 Nano 30B A3B costs less

Agent loop

200K input + 10K output tokens

Gemma 3 12B$0.017Exceeds the context window
Nemotron 3 Nano 30B A3B$0.013Fits in one request

Nemotron 3 Nano 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 Gemma 3 12B and Nemotron 3 Nano 30B A3B
SpecificationGoogle Gemma 3 12BNVIDIA Nemotron 3 Nano 30B A3B
MakerGoogleNVIDIA
WeightsOpen weightsOpen weights
ReleasedMarch 13, 2025December 14, 2025
Context window131K262K
Input $/Mtok$0.075$0.055
Output $/Mtok$0.1625$0.22
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemma 3 12B better than Nemotron 3 Nano 30B A3B for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Gemma 3 12B for output tokens 1.4x cheaper ($0.1625 vs $0.22 per million) and work measured by AA Coding Index or LiveCodeBench (Artificial Analysis run), where it leads.

Which is cheaper, Gemma 3 12B or Nemotron 3 Nano 30B A3B?

Gemma 3 12B costs $0.075 per million input tokens and $0.1625 per million output tokens; Nemotron 3 Nano 30B A3B costs $0.055 and $0.22. Cost per solved task is known for neither of them.

Which has the bigger context window, Gemma 3 12B or Nemotron 3 Nano 30B A3B?

Nemotron 3 Nano 30B A3B, with 262K tokens against 131K.

How many benchmarks is the Gemma 3 12B vs Nemotron 3 Nano 30B A3B comparison based on?

36 benchmarks have a result for both models as of October 22, 2026. Gemma 3 12B's coding index rests on 0 independent benchmarks and Nemotron 3 Nano 30B A3B's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next