Enter opens the highlighted result. Esc closes.

Nova Micro 1.0 vs Gemma 3 12B

Nova Micro 1.0 leads Gemma 3 12B on the Artificial Analysis Intelligence Index, 5.9 to 3.8, at about the same output price ($0.14 vs $0.1625 per million). Updated October 21, 2026; 33 benchmarks have a result for both.

Replace Nova Micro 1.0 Replace Gemma 3 12B
Nova Micro 1.0Amazon
Coding
No data
Overall
No data
Output $/M
$0.14
7
Gemma 3 12BGoogle
Coding
No data
Overall
28.0
Output $/M
$0.1625
26
vs boards won of 33 shared

Nova Micro 1.0 or Gemma 3 12B: which to pick

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

Pick Nova Micro 1.0 for output tokens 1.2x cheaper ($0.14 vs $0.1625 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Model ANova Micro 1.0AmazonNot benchmarkedNo coding benchmarkNo cost per solved taskModel BGemma 3 12BGoogleNot benchmarkedNo coding benchmarkNo cost per solved task
Shared results
33
Nova Micro 1.0 only
6
Gemma 3 12B only
10
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
Nova Micro 1.0 not benchmarked, Gemma 3 12B not benchmarked.
Widest category
Knowledge, 22.5 points apart on average over 5 shared benchmarks.
Widest single gap
LMArena Text: business and finance: Nova Micro 1.0 1182 against Gemma 3 12B 1376, 32.4 points apart on a 0 to 100 scale.
Output price
Nova Micro 1.0 $0.14 against Gemma 3 12B $0.1625 per million output tokens, 1.2x apart.
Context window
Nova Micro 1.0 takes 128K tokens, Gemma 3 12B 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
CategoryNova Micro 1.0BarsGemma 3 12BTies
Coding22
Intelligence10
Agentic work10
Reasoning23
Math04
Knowledge05
General112
All shared boards726
CodingLike-for-likeSix of these boards feed the Vibe Coding Index22
Coding benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
AA Coding IndexNot measured5.8Only Gemma 3 12B measured
LiveCodeBench (Artificial Analysis run)14.0%13.7%Nova Micro 1.0
SciCodeNot measured16.4%Only Gemma 3 12B measured
Terminal-Bench Hard (Artificial Analysis run)1.5%0.8%Nova Micro 1.0
LMArena Text: coding12181281Gemma 3 12B
LMArena Text: software and IT12151313Gemma 3 12B
SciCodeNot measured0.0%Only Gemma 3 12B measured
Neither model has a result on the other 32 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
AA Intelligence Index5.93.8Nova Micro 1.0
Epoch Capabilities IndexNot measured123.5Only Gemma 3 12B measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
AA Agentic IndexNot measured0.1Only Gemma 3 12B measured
τ²-Bench Telecom14.0%10.8%Nova Micro 1.0
Berkeley Function Calling Leaderboard (BFCL)Not measured30.4%Only Gemma 3 12B measured
Neither model has a result on the other 38 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index23
Reasoning benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
GPQA Diamond (Artificial Analysis run)35.8%34.9%Nova Micro 1.0
Humanity's Last Exam (Artificial Analysis run)4.6%4.2%Nova Micro 1.0
GPQA Diamond (Epoch AI run)Not measured39.5%Only Gemma 3 12B measured
LMArena Text: hard prompts11911309Gemma 3 12B
LMArena Text: expert11831248Gemma 3 12B
LMArena Text: hard prompts in English12031318Gemma 3 12B
Neither model has a result on the other 16 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index04
Math benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
AA Math Index6.018.3Gemma 3 12B
AIME 2025 (Artificial Analysis run)6.0%18.3%Gemma 3 12B
OTIS Mock AIMENot measured16.7%Only Gemma 3 12B measured
LMArena Text: math12061307Gemma 3 12B
LMArena Text: mathematical work12121336Gemma 3 12B
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
Knowledge benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
MMLU Pro (Artificial Analysis run)53.1%59.5%Gemma 3 12B
MMLU70.8%Not measuredOnly Nova Micro 1.0 measured
LMArena Text: science12101366Gemma 3 12B
LMArena Text: medicine and healthcare12071334Gemma 3 12B
LMArena Text: legal and government12211381Gemma 3 12B
LMArena Text: business and finance11821376Gemma 3 12B
Neither model has a result on the other 12 boards here
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Web design benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
Neither model has a result on 9 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index112
General benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
IFBench29.4%36.7%Gemma 3 12B
Long context reasoning (AA-LCR)13.0%8.3%Nova Micro 1.0
DTBenchNot measured48.8%Only Gemma 3 12B measured
LMCANot measured4.5%Only Gemma 3 12B measured
LMArena Text12081334Gemma 3 12B
LMArena Text: instruction following11741299Gemma 3 12B
LMArena Text: creative writing11721331Gemma 3 12B
LMArena Text: English12241349Gemma 3 12B
LMArena Text: non-English11861318Gemma 3 12B
LMArena Text: Chinese1209Not measuredOnly Nova Micro 1.0 measured
LMArena Text: German11921370Gemma 3 12B
LMArena Text: Japanese1154Not measuredOnly Nova Micro 1.0 measured
LMArena Text: Korean1150Not measuredOnly Nova Micro 1.0 measured
LMArena Text: French1238Not measuredOnly Nova Micro 1.0 measured
LMArena Text: Spanish1225Not measuredOnly Nova Micro 1.0 measured
LMArena Text: Russian11851335Gemma 3 12B
LMArena Text: multi-turn11781334Gemma 3 12B
LMArena Text: longer queries12051317Gemma 3 12B
LMArena Text: writing and language11921307Gemma 3 12B
LMArena Text: entertainment and media11631297Gemma 3 12B
Neither model has a result on the other 26 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Nova Micro 1.0 and Gemma 3 12B
BenchmarkNova Micro 1.0Gemma 3 12BLeader
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

Nova Micro 1.0$0.0001Fits in one request
Gemma 3 12B$0.0002Fits in one request

Nova Micro 1.0 costs less

Repository review

50K input + 3K output tokens

Nova Micro 1.0$0.0022Fits in one request
Gemma 3 12B$0.0042Fits in one request

Nova Micro 1.0 costs less

Agent loop

200K input + 10K output tokens

Nova Micro 1.0$0.0084Exceeds the context window
Gemma 3 12B$0.017Exceeds the context window

Nova Micro 1.0 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 Nova Micro 1.0 and Gemma 3 12B
SpecificationAmazon Nova Micro 1.0Google Gemma 3 12B
MakerAmazonGoogle
WeightsNo public weights listedOpen weights
ReleasedDecember 5, 2024March 13, 2025
Context window128K131K
Input $/Mtok$0.035$0.075
Output $/Mtok$0.14$0.1625
Coding indexNot benchmarkedNot benchmarked
Independent measurements behind it00
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Nova Micro 1.0 better than Gemma 3 12B for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Nova Micro 1.0 for output tokens 1.2x cheaper ($0.14 vs $0.1625 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.

Which is cheaper, Nova Micro 1.0 or Gemma 3 12B?

Nova Micro 1.0 costs $0.035 per million input tokens and $0.14 per million output tokens; Gemma 3 12B costs $0.075 and $0.1625. Cost per solved task is known for neither of them.

Which has the bigger context window, Nova Micro 1.0 or Gemma 3 12B?

Gemma 3 12B, with 131K tokens against 128K.

How many benchmarks is the Nova Micro 1.0 vs Gemma 3 12B comparison based on?

33 benchmarks have a result for both models as of October 21, 2026. Nova Micro 1.0's coding index rests on 0 independent benchmarks and Gemma 3 12B's on 0. Where only one model has a result, the page says so instead of naming a winner.

Compare next