Model comparison

Gemma 3 12B vs GPT-5-Codex

GPT-5-Codex is the stronger model overall, scoring 37.3 to 31.8 on the Noometry Index. Gemma 3 12B costs 46× less per token, which makes it the better buy when GPT-5-Codex's lead doesn't matter for your workload.

Last verified October 10, 2026. 0 shared benchmarks.

Gemma 3 12B

Google

31.8

Rank #255 Confirmed

GPT-5-Codex OpenAI

37.3

Rank #190 Reported

Summary

The widest gap is in reasoning, where GPT-5-Codex leads 30.0 to 15.2.
Gemma 3 12B is cheaper at $0.05 / $0.15 per million input/output tokens, against $1.25 / $10 for GPT-5-Codex.
GPT-5-Codex accepts more context: 400K tokens versus 131K.
Gemma 3 12B has downloadable open weights; the other is API-only.

Side by side

Gemma 3 12B and GPT-5-Codex specifications
	Gemma 3 12B	GPT-5-Codex
Provider	Google	OpenAI
Noometry Index	31.8	37.3
Released	2025-03-12	2025-09-15
Weights	Open	Proprietary
Context window	131K	400K
Max output	8K	128K
Input $ / M tokens	$0.05	$1.25
Output $ / M tokens	$0.15	$10
Results tracked	24	3

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

Category by category

Coding GPT-5-Codex leads

Gemma 3 12B: 31.2 (#276), GPT-5-Codex: 41.9 (#100)

Coding benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
SciCode	17.4%	—
WeirdML	—	54.5%
LMArena Coding	1281	—

Agentic & Tool Use GPT-5-Codex leads

Gemma 3 12B: 25.3 (#101), GPT-5-Codex: 30.8 (#66)

Agentic & Tool Use benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
Terminal-Bench	—	44.3%
Berkeley Function Calling Leaderboard	30.4%	—

Reasoning GPT-5-Codex leads

Gemma 3 12B: 15.2 (#307), GPT-5-Codex: 30.0 (#80)

Reasoning benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
Kagi LLM Benchmark	—	70.3%
CritPt	0%	—
Chess Puzzles	0%	—
LMArena Hard Prompts	1309	—
DTBench	48.8%	—
LMCA	4.5%	—
Epoch Capabilities Index	123.49	—

Math Not comparable

Gemma 3 12B: 21.0 (#273), GPT-5-Codex: —

Math benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
OTIS Mock AIME 2024-2025	16.7%	—
LMArena Math	1307	—

Knowledge Not comparable

Gemma 3 12B: 26.5 (#250), GPT-5-Codex: —

Knowledge benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
GPQA Diamond	39.5%	—
Vectara Hallucination Rate	4.4%	—
LMArena Expert	1248	—

Multimodal Not comparable

Gemma 3 12B: —, GPT-5-Codex: —

Multimodal benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
MindCube	46.7%	—

Multilingual Not comparable

Gemma 3 12B: 45.9 (#160), GPT-5-Codex: —

Multilingual benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
LMArena Non-English	1318	—
LMArena German	1370	—
LMArena Russian	1335	—

Instruction Following Not comparable

Gemma 3 12B: 68.8 (#179), GPT-5-Codex: —

Instruction Following benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
LMArena Instruction Following	1299	—

Long Context Not comparable

Gemma 3 12B: 40.0 (#157), GPT-5-Codex: —

Long Context benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
LMArena Longer Query	1317	—

Writing & Preference Not comparable

Gemma 3 12B: 47.7 (#205), GPT-5-Codex: —

Writing & Preference benchmarks
Benchmark	Gemma 3 12B	GPT-5-Codex
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 GPT-5-Codex?

Which is cheaper, Gemma 3 12B or GPT-5-Codex?

Gemma 3 12B is cheaper. It lists at $0.05 per million input tokens and $0.15 per million output tokens; GPT-5-Codex lists at $1.25 and $10.

Is Gemma 3 12B or GPT-5-Codex better for coding?

GPT-5-Codex scores higher on coding benchmarks: 41.9 versus 31.2 in the Noometry coding category.

Which has the bigger context window?

GPT-5-Codex does, with 400K tokens against 131K.

How many benchmarks do Gemma 3 12B and GPT-5-Codex share?

0 benchmarks have published results for both models. Gemma 3 12B has 24 scored results on Noometry and GPT-5-Codex has 3.