Model comparison

Gemma 3 4B vs Gemma 4 26B A4B IT

Gemma 4 26B A4B IT is the stronger model overall, scoring 43.5 to 28.1 on the Noometry Index. Gemma 3 4B costs 2.9× less per token, which makes it the better buy when Gemma 4 26B A4B IT's lead doesn't matter for your workload.

Last verified October 12, 2026. 19 shared benchmarks.

Gemma 3 4B

Google

28.1

Rank #326 Confirmed

Gemma 4 26B A4B IT

Google

43.5

Rank #92 Confirmed

Summary

They share 19 benchmarks with published results for both. Gemma 3 4B scores higher in 0 categories and Gemma 4 26B A4B IT in 8 categories; 8 gaps are clear of the uncertainty.
The widest gap is in knowledge, where Gemma 4 26B A4B IT leads 45.7 to 11.8.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 7.5% for Gemma 3 4B and 82.2% for Gemma 4 26B A4B IT.
Gemma 3 4B is cheaper at $0.04 / $0.08 per million input/output tokens, against $0.09 / $0.30 for Gemma 4 26B A4B IT.
Gemma 4 26B A4B IT accepts more context: 262K tokens versus 131K.

Side by side

Gemma 3 4B and Gemma 4 26B A4B IT specifications
	Gemma 3 4B	Gemma 4 26B A4B IT
Provider	Google	Google
Noometry Index	28.1	43.5
Released	2025-03-12	2026-04-02
Weights	Open	Open
Context window	131K	262K
Max output	4K	33K
Input $ / M tokens	$0.04	$0.09
Output $ / M tokens	$0.08	$0.30
Results tracked	22	28

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

Category by category

Coding Gemma 4 26B A4B IT leads

Gemma 3 4B: 35.9 (#215), Gemma 4 26B A4B IT: 39.0 (#164)

Coding benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
LMArena Coding	1230	1447
LMArena WebDev	—	1359
SciCode	—	40%
WeirdML	—	35.2%
ALE-Bench	—	927.17

Agentic & Tool Use Not comparable

Gemma 3 4B: 20.9 (#142), Gemma 4 26B A4B IT: —

Agentic & Tool Use benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
Berkeley Function Calling Leaderboard	19.6%	—

Reasoning Gemma 4 26B A4B IT leads

Gemma 3 4B: 13.2 (#335), Gemma 4 26B A4B IT: 21.8 (#213)

Reasoning benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
Chess Puzzles	0%	6%
LMArena Hard Prompts	1253	1439
DTBench	50.9%	74.9%
LMCA	2.8%	29.7%
Epoch Capabilities Index	116.02	141.85
Kagi LLM Benchmark	25.2%	—
CritPt	—	0%

Math Gemma 4 26B A4B IT leads

Gemma 3 4B: 16.8 (#292), Gemma 4 26B A4B IT: 47.6 (#67)

Math benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
OTIS Mock AIME 2024-2025	7.5%	82.2%
LMArena Math	1239	1470

Knowledge Gemma 4 26B A4B IT leads

Gemma 3 4B: 11.8 (#299), Gemma 4 26B A4B IT: 45.7 (#85)

Knowledge benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
GPQA Diamond	23.2%	73.2%
Vectara Hallucination Rate	6.4%	5.2%
LMArena Expert	1223	1447

Multimodal Not comparable

Gemma 3 4B: —, Gemma 4 26B A4B IT: 40.6 (#46)

Multimodal benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
LMArena Vision	—	1260

Multilingual Gemma 4 26B A4B IT leads

Gemma 3 4B: 42.5 (#194), Gemma 4 26B A4B IT: 53.1 (#71)

Multilingual benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
LMArena Non-English	1273	1421
LMArena Russian	1294	1434
LMArena Chinese	—	1495
LMArena French	—	1460
LMArena German	1281	—
LMArena Spanish	—	1417

Instruction Following Gemma 4 26B A4B IT leads

Gemma 3 4B: 65.2 (#225), Gemma 4 26B A4B IT: 74.8 (#82)

Instruction Following benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
LMArena Instruction Following	1239	1420

Long Context Gemma 4 26B A4B IT leads

Gemma 3 4B: 38.7 (#194), Gemma 4 26B A4B IT: 43.6 (#91)

Long Context benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
LMArena Longer Query	1273	1428

Writing & Preference Gemma 4 26B A4B IT leads

Gemma 3 4B: 42.0 (#239), Gemma 4 26B A4B IT: 58.6 (#115)

Writing & Preference benchmarks
Benchmark	Gemma 3 4B	Gemma 4 26B A4B IT
LMArena Text	1291	1434
LMArena Creative Writing	1271	1402
EQ-Bench Creative Writing	1068	1305
LMArena Multi-Turn	1255	1441

Frequently asked questions

Is Gemma 3 4B better than Gemma 4 26B A4B IT?

Which is cheaper, Gemma 3 4B or Gemma 4 26B A4B IT?

Gemma 3 4B is cheaper. It lists at $0.04 per million input tokens and $0.08 per million output tokens; Gemma 4 26B A4B IT lists at $0.09 and $0.30.

Is Gemma 3 4B or Gemma 4 26B A4B IT better for coding?

Gemma 4 26B A4B IT scores higher on coding benchmarks: 39.0 versus 35.9 in the Noometry coding category.

Which has the bigger context window?

Gemma 4 26B A4B IT does, with 262K tokens against 131K.

How many benchmarks do Gemma 3 4B and Gemma 4 26B A4B IT share?

19 benchmarks have published results for both models. Gemma 3 4B has 22 scored results on Noometry and Gemma 4 26B A4B IT has 28.