Model comparison

Gemma 3 4B vs Nova Premier 1.0

Nova Premier 1.0 is the stronger model overall, scoring 38.3 to 28.1 on the Noometry Index. Gemma 3 4B costs 100× less per token, which makes it the better buy when Nova Premier 1.0's lead doesn't matter for your workload.

Last verified October 10, 2026. 1 shared benchmarks.

Gemma 3 4B

Google

28.1

Rank #326 Confirmed

Nova Premier 1.0 Amazon

38.3

Rank #189 Confirmed

Summary

They share 1 benchmark with published results for both. Gemma 3 4B scores higher in 0 categories and Nova Premier 1.0 in 5 categories; 5 gaps are clear of the uncertainty.
The widest gap is in knowledge, where Nova Premier 1.0 leads 35.8 to 11.8.
The biggest single-benchmark swing is Kagi LLM Benchmark: 25.2% for Gemma 3 4B and 44.8% for Nova Premier 1.0.
Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $2.50 / $12.50 for Nova Premier 1.0.
Nova Premier 1.0 accepts more context: 1M tokens versus 131K.
Gemma 3 4B has downloadable open weights; the other is API-only.

Side by side

Gemma 3 4B and Nova Premier 1.0 specifications
	Gemma 3 4B	Nova Premier 1.0
Provider	Google	Amazon
Noometry Index	28.1	38.3
Released	2025-03-12	2025-04-30
Weights	Open	Proprietary
Context window	131K	1M
Max output	4K	10K
Input $ / M tokens	$0.04	$2.50
Output $ / M tokens	$0.08	$12.50
Results tracked	22	6

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Not comparable

Gemma 3 4B: 35.9 (#215), Nova Premier 1.0: —

Coding benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
LMArena Coding	1230	—

Agentic & Tool Use Not comparable

Gemma 3 4B: 20.9 (#142), Nova Premier 1.0: —

Agentic & Tool Use benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
Berkeley Function Calling Leaderboard	19.6%	—

Reasoning Nova Premier 1.0 leads

Gemma 3 4B: 13.2 (#335), Nova Premier 1.0: 23.0 (#185)

Reasoning benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
Kagi LLM Benchmark	25.2%	44.8%
Chess Puzzles	0%	—
LMArena Hard Prompts	1253	—
DTBench	50.9%	—
LMCA	2.8%	—
Epoch Capabilities Index	116.02	—

Math Nova Premier 1.0 leads

Gemma 3 4B: 16.8 (#292), Nova Premier 1.0: 33.8 (#199)

Math benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
OTIS Mock AIME 2024-2025	7.5%	—
Omni-MATH	—	35%
LMArena Math	1239	—

Knowledge Nova Premier 1.0 leads

Gemma 3 4B: 11.8 (#299), Nova Premier 1.0: 35.8 (#180)

Knowledge benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
GPQA Diamond	23.2%	—
MMLU-Pro	—	72.6%
Vectara Hallucination Rate	6.4%	—
GPQA (HELM)	—	51.8%
LMArena Expert	1223	—

Multilingual Not comparable

Gemma 3 4B: 42.5 (#194), Nova Premier 1.0: —

Multilingual benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
LMArena Non-English	1273	—
LMArena German	1281	—
LMArena Russian	1294	—

Instruction Following Nova Premier 1.0 leads

Gemma 3 4B: 65.2 (#225), Nova Premier 1.0: 66.8 (#204)

Instruction Following benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
IFEval	—	80.3%
LMArena Instruction Following	1239	—

Long Context Not comparable

Gemma 3 4B: 38.7 (#194), Nova Premier 1.0: —

Long Context benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
LMArena Longer Query	1273	—

Writing & Preference Nova Premier 1.0 leads

Gemma 3 4B: 42.0 (#239), Nova Premier 1.0: 51.0 (#178)

Writing & Preference benchmarks
Benchmark	Gemma 3 4B	Nova Premier 1.0
LMArena Text	1291	—
LMArena Creative Writing	1271	—
EQ-Bench Creative Writing	1068	—
WildBench	—	78.8%
LMArena Multi-Turn	1255	—

Frequently asked questions

Is Gemma 3 4B better than Nova Premier 1.0?

Which is cheaper, Gemma 3 4B or Nova Premier 1.0?

Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Nova Premier 1.0 lists at $2.50 and $12.50.

Which has the bigger context window?

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

How many benchmarks do Gemma 3 4B and Nova Premier 1.0 share?

1 benchmark has published results for both models. Gemma 3 4B has 22 scored results on Noometry and Nova Premier 1.0 has 6.