Model comparison

Gemma 3 12B vs Phi-4 Mini

Gemma 3 12B is the stronger model overall, scoring 32.1 to 30.9 on the Noometry Index.

Last verified October 11, 2026. 3 shared benchmarks.

Gemma 3 12B

Google

32.1

Rank #262 Confirmed

Phi-4 Mini

Microsoft

30.9

Rank #283 Reported

Summary

They share 3 benchmarks with published results for both. Gemma 3 12B scores higher in 2 categories and Phi-4 Mini in 1 category; 3 gaps are clear of the uncertainty.
The widest gap is in reasoning, where Phi-4 Mini leads 22.4 to 15.7.
The biggest single-benchmark swing is Vectara Hallucination Rate: 4.4% for Gemma 3 12B and 23.5% for Phi-4 Mini.
Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $0.075 / $0.30 for Phi-4 Mini.
Gemma 3 12B accepts more context: 131K tokens versus 128K.

Side by side

Gemma 3 12B and Phi-4 Mini specifications
	Gemma 3 12B	Phi-4 Mini
Provider	Google	Microsoft
Noometry Index	32.1	30.9
Released	2025-03-12	2024-12-11
Weights	Open	Open
Context window	131K	128K
Max output	8K	4K
Input $ / M tokens	$0.05	$0.075
Output $ / M tokens	$0.15	$0.30
Results tracked	24	3

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

Category by category

Coding Gemma 3 12B leads

Gemma 3 12B: 31.7 (#280), Phi-4 Mini: 28.1 (#317)

Coding benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
SciCode	17.4%	10.8%
LMArena Coding	1281	—

Agentic & Tool Use Not comparable

Gemma 3 12B: 25.5 (#108), Phi-4 Mini: —

Agentic & Tool Use benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
Berkeley Function Calling Leaderboard	30.4%	—

Reasoning Phi-4 Mini leads

Gemma 3 12B: 15.7 (#313), Phi-4 Mini: 22.4 (#195)

Reasoning benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
CritPt	0%	0%
Chess Puzzles	0%	—
LMArena Hard Prompts	1309	—
DTBench	48.8%	—
LMCA	4.5%	—
Epoch Capabilities Index	123.5	—

Math Not comparable

Gemma 3 12B: 22.3 (#279), Phi-4 Mini: —

Math benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
OTIS Mock AIME 2024-2025	16.7%	—
LMArena Math	1307	—

Knowledge Gemma 3 12B leads

Gemma 3 12B: 26.5 (#257), Phi-4 Mini: 25.3 (#262)

Knowledge benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
Vectara Hallucination Rate	4.4%	23.5%
GPQA Diamond	39.5%	—
LMArena Expert	1248	—

Multimodal Not comparable

Gemma 3 12B: —, Phi-4 Mini: —

Multimodal benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
MindCube	46.7%	—

Multilingual Not comparable

Gemma 3 12B: 45.7 (#165), Phi-4 Mini: —

Multilingual benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
LMArena Non-English	1318	—
LMArena German	1370	—
LMArena Russian	1335	—

Instruction Following Not comparable

Gemma 3 12B: 68.6 (#186), Phi-4 Mini: —

Instruction Following benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
LMArena Instruction Following	1299	—

Long Context Not comparable

Gemma 3 12B: 40.0 (#162), Phi-4 Mini: —

Long Context benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
LMArena Longer Query	1317	—

Writing & Preference Not comparable

Gemma 3 12B: 47.5 (#209), Phi-4 Mini: —

Writing & Preference benchmarks
Benchmark	Gemma 3 12B	Phi-4 Mini
LMArena Text	1334	—
LMArena Creative Writing	1331	—
EQ-Bench Creative Writing	1126	—
LMArena Multi-Turn	1334	—

Frequently asked questions

Is Gemma 3 12B better than Phi-4 Mini?

Gemma 3 12B is the stronger model overall, scoring 32.1 to 30.9 on the Noometry Index.

Which is cheaper, Gemma 3 12B or Phi-4 Mini?

Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; Phi-4 Mini lists at $0.075 and $0.30.

Is Gemma 3 12B or Phi-4 Mini better for coding?

Gemma 3 12B scores higher on coding benchmarks: 31.7 versus 28.1 in the Noometry coding category.

Which has the bigger context window?

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

How many benchmarks do Gemma 3 12B and Phi-4 Mini share?

3 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and Phi-4 Mini has 3.