all the models — AI benchmark observatory
← Models

Gemma 3 12B

Google · open weight · gemma-3-12b-it

Gemma 3 12B is a 12-billion-parameter vision-language model from Google, handling text and image input and generating text output. It features a 128K context window, multilingual support, and open weights. Suitable for question answering, summarization, reasoning, and image understanding tasks.

GSM8kIFEvalDocVQAHumanEvalMATHMMLU-Pro

Benchmark scores

BenchmarkScore
GSM8k0.94
IFEval0.89
DocVQA0.87
HumanEval0.85
MATH0.84
MMLU-Pro0.61
GPQA0.41
LiveCodeBench0.25

Pricing

  • DeepInfra$0.05 / $0.15

Input / output per 1M tokens

AA metrics

No Artificial Analysis link yet.

Arena Elo

  • No Arena snapshot linked.
Gemma 3 12B Benchmarks · all the models