Nova Premier 1.0 vs Gemma 3 27B

Nova Premier 1.0 leads Gemma 3 27B on the Artificial Analysis Intelligence Index, 9.2 to 4.9, and Gemma 3 27B costs 41.7x less per output token ($0.30 vs $12.50 per million). Updated October 21, 2026; 11 benchmarks have a result for both.

Replace Nova Premier 1.0 Replace Gemma 3 27B

Nova Premier 1.0Amazon

Coding: No data
Overall: 9.2
Output $/M: $12.50

Gemma 3 27BGoogle

Coding: No data
Overall: 4.9
Output $/M: $0.30

vs boards won of 11 shared

Nova Premier 1.0 or Gemma 3 27B: which to pick

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

Pick Nova Premier 1.0 for a larger context window (1M against 131K tokens) and work measured by Humanity's Last Exam (Artificial Analysis run) or AA Math Index, where it leads.

Model ANova Premier 1.0AmazonNot benchmarkedNo coding benchmarkNo cost per solved task Model BGemma 3 27BGoogleNot benchmarkedNo coding benchmarkNo cost per solved task

11251

Shared results: 11
Nova Premier 1.0 only: 2
Gemma 3 27B only: 51
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.

Coding
Vibe Coding Index, six public coding benchmarks
No call
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
Nova Premier 1.0
Nova Premier 1.0 leads, 9.2 against 4.9.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Gemma 3 27B has a AA Agentic Index result, 0.1.
Directional only
Web design
Design Arena: websites
No call
Only Nova Premier 1.0 has a Design Arena: websites result, 835.
Directional only
Long documents
Listed context window
Nova Premier 1.0
Nova Premier 1.0 takes 1M tokens against 131K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Gemma 3 27B
Nova Premier 1.0 $0.0088, Gemma 3 27B $0.0002 per turn.
List prices

At a glance

Coding index: Nova Premier 1.0 not benchmarked, Gemma 3 27B not benchmarked.
Widest category: Agentic work, 27.8 points apart on average over 1 shared benchmark.
Widest single gap: τ²-Bench Telecom: Nova Premier 1.0 38.3% against Gemma 3 27B 10.5%, 27.8 points apart on a 0 to 100 scale.
Output price: Nova Premier 1.0 $12.50 against Gemma 3 27B $0.30 per million output tokens, 41.7x apart.
Context window: Nova Premier 1.0 takes 1M tokens, Gemma 3 27B 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
Category	Nova Premier 1.0	Gemma 3 27B
Coding	2	0
Intelligence	1	0
Agentic work	1	0
Reasoning	1	1
Math	0	2
Knowledge	1	0
General	2	0
All shared boards	8	3

CodingLike-for-likeSix of these boards feed the Vibe Coding Index20

Coding benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
Aider polyglot	Not measured	4.9%	Only Gemma 3 27B measured
AA Coding Index	Not measured	10.1	Only Gemma 3 27B measured
LiveCodeBench (Artificial Analysis run)	31.7%	13.7%	Nova Premier 1.0
SciCode	Not measured	23.3%	Only Gemma 3 27B measured
Terminal-Bench Hard (Artificial Analysis run)	6.8%	3.8%	Nova Premier 1.0
ALE-Bench	147.4	Not measured	Only Nova Premier 1.0 measured
LMArena Text: coding	Not measured	1322	Only Gemma 3 27B measured
LMArena Text: software and IT	Not measured	1354	Only Gemma 3 27B measured
Neither model has a result on the other 30 boards here

IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10

Intelligence benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
AA Intelligence Index	9.2	4.9	Nova Premier 1.0
Epoch Capabilities Index	Not measured	130.0	Only Gemma 3 27B measured

Agentic workLike-for-likePublic boards, shown as published, kept out of the index10

Agentic work benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
AA Agentic Index	Not measured	0.1	Only Gemma 3 27B measured
τ²-Bench Telecom	38.3%	10.5%	Nova Premier 1.0
Neither model has a result on the other 37 boards here

ReasoningLike-for-likePublic boards, shown as published, kept out of the index11

Reasoning benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
GPQA Diamond (Artificial Analysis run)	56.9%	42.8%	Nova Premier 1.0
Humanity's Last Exam (Artificial Analysis run)	4.2%	4.4%	Gemma 3 27B
GPQA Diamond (Epoch AI run)	Not measured	47.7%	Only Gemma 3 27B measured
LMArena Text: hard prompts	Not measured	1340	Only Gemma 3 27B measured
LMArena Text: expert	Not measured	1304	Only Gemma 3 27B measured
LMArena Text: hard prompts in English	Not measured	1353	Only Gemma 3 27B measured
LMArena Vision: diagrams	Not measured	1156	Only Gemma 3 27B measured
LMArena Vision: homework	Not measured	1160	Only Gemma 3 27B measured
Neither model has a result on the other 13 boards here

MathLike-for-likePublic boards, shown as published, kept out of the index02

Math benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
AA Math Index	17.3	20.7	Gemma 3 27B
AIME 2025 (Artificial Analysis run)	17.3%	20.7%	Gemma 3 27B
OTIS Mock AIME	Not measured	22.5%	Only Gemma 3 27B measured
MATH level 5	Not measured	74.0%	Only Gemma 3 27B measured
LMArena Text: math	Not measured	1312	Only Gemma 3 27B measured
LMArena Text: mathematical work	Not measured	1327	Only Gemma 3 27B measured
Neither model has a result on the other 7 boards here

KnowledgeLike-for-likePublic boards, shown as published, kept out of the index10

Knowledge benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
MMLU Pro (Artificial Analysis run)	73.3%	66.9%	Nova Premier 1.0
GeoBench	Not measured	52.0%	Only Gemma 3 27B measured
LMArena Text: science	Not measured	1383	Only Gemma 3 27B measured
LMArena Text: medicine and healthcare	Not measured	1391	Only Gemma 3 27B measured
LMArena Text: legal and government	Not measured	1375	Only Gemma 3 27B measured
LMArena Text: business and finance	Not measured	1360	Only Gemma 3 27B measured
Neither model has a result on the other 11 boards here

Web designDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared

Web design benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
Design Arena: websites	835	Not measured	Only Nova Premier 1.0 measured
Neither model has a result on the other 8 boards here

GeneralLike-for-likePublic boards, shown as published, kept out of the index20

General benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
IFBench	36.2%	31.8%	Nova Premier 1.0
Long context reasoning (AA-LCR)	34.3%	7.3%	Nova Premier 1.0
DTBench	Not measured	52.5%	Only Gemma 3 27B measured
LMCA	Not measured	12.3%	Only Gemma 3 27B measured
Fiction.LiveBench	Not measured	33.3%	Only Gemma 3 27B measured
Creative writing (Lech Mazur)	Not measured	79.9%	Only Gemma 3 27B measured
LMArena Text	Not measured	1358	Only Gemma 3 27B measured
LMArena Text: instruction following	Not measured	1321	Only Gemma 3 27B measured
LMArena Text: creative writing	Not measured	1345	Only Gemma 3 27B measured
LMArena Vision	Not measured	1165	Only Gemma 3 27B measured
LMArena Text: English	Not measured	1377	Only Gemma 3 27B measured
LMArena Text: non-English	Not measured	1334	Only Gemma 3 27B measured
LMArena Text: Chinese	Not measured	1346	Only Gemma 3 27B measured
LMArena Text: German	Not measured	1362	Only Gemma 3 27B measured
LMArena Text: Japanese	Not measured	1287	Only Gemma 3 27B measured
LMArena Text: Korean	Not measured	1308	Only Gemma 3 27B measured
LMArena Text: French	Not measured	1368	Only Gemma 3 27B measured
LMArena Text: Spanish	Not measured	1349	Only Gemma 3 27B measured
LMArena Text: Russian	Not measured	1349	Only Gemma 3 27B measured
LMArena Text: Polish	Not measured	1345	Only Gemma 3 27B measured
LMArena Text: multi-turn	Not measured	1345	Only Gemma 3 27B measured
LMArena Text: longer queries	Not measured	1333	Only Gemma 3 27B measured
LMArena Text: writing and language	Not measured	1347	Only Gemma 3 27B measured
LMArena Text: entertainment and media	Not measured	1320	Only Gemma 3 27B measured
LMArena Vision: English	Not measured	1188	Only Gemma 3 27B measured
LMArena Vision: Chinese	Not measured	1164	Only Gemma 3 27B measured
LMArena Vision: reading text	Not measured	1156	Only Gemma 3 27B measured
LMArena Vision: humor	Not measured	1166	Only Gemma 3 27B measured
LMArena Vision: captioning	Not measured	1151	Only Gemma 3 27B measured
LMArena Vision: recognizing things	Not measured	1196	Only Gemma 3 27B measured
LMArena Vision: creative writing	Not measured	1177	Only Gemma 3 27B measured
Neither model has a result on the other 15 boards here

Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared

Image generation benchmarks for Nova Premier 1.0 and Gemma 3 27B
Benchmark	Nova Premier 1.0	Gemma 3 27B	Leader
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 Premier 1.0$0.0088Fits in one request

Gemma 3 27B$0.0002Fits in one request

Gemma 3 27B costs less

Repository review

50K input + 3K output tokens

Nova Premier 1.0$0.163Fits in one request

Gemma 3 27B$0.0049Fits in one request

Gemma 3 27B costs less

Agent loop

200K input + 10K output tokens

Nova Premier 1.0$0.625Fits in one request

Gemma 3 27B$0.019Exceeds the context window

Gemma 3 27B 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 Premier 1.0 and Gemma 3 27B
Specification	Amazon Nova Premier 1.0	Google Gemma 3 27B
Maker	Amazon	Google
Weights	No public weights listed	Open weights
Released	October 31, 2025	March 12, 2025
Context window	1M	131K
Input $/Mtok	$2.50	$0.08
Output $/Mtok	$12.50	$0.30
Coding index	Not benchmarked	Not benchmarked
Independent measurements behind it	0	0
Cost per solved task	Not computable	Not computable
Source	OpenRouter listing	OpenRouter listing

Questions

Is Nova Premier 1.0 better than Gemma 3 27B for coding?

Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Nova Premier 1.0 for a larger context window (1M against 131K tokens) and work measured by Humanity's Last Exam (Artificial Analysis run) or AA Math Index, where it leads.

Which is cheaper, Nova Premier 1.0 or Gemma 3 27B?

Nova Premier 1.0 costs $2.50 per million input tokens and $12.50 per million output tokens; Gemma 3 27B costs $0.08 and $0.30. Cost per solved task is known for neither of them.

Which has the bigger context window, Nova Premier 1.0 or Gemma 3 27B?

Nova Premier 1.0, with 1M tokens against 131K.

How many benchmarks is the Nova Premier 1.0 vs Gemma 3 27B comparison based on?

11 benchmarks have a result for both models as of October 21, 2026. Nova Premier 1.0's coding index rests on 0 independent benchmarks and Gemma 3 27B's on 0. Where only one model has a result, the page says so instead of naming a winner.

Nova Premier 1.0 or Gemma 3 27B: which to pick

Which one for your work

Coding

Intelligence

Agentic work

Web design

Long documents

Everyday cost

At a glance

Benchmark by benchmark

What each workload costs

Chat turn

Repository review

Agent loop

Specification differences

Questions

Compare next