Google

Gemma 4 12B

Jun 3, 2026Open weights

Compare

Intelligence

9.4

#446 of 744Rank 446 of 744 measured models

Output speed

111 t/s

#157 of 403Rank 157 of 403 measured models

Price/1M

$0.15

#128 of 659Rank 128 of 659 measured models

Context window

262K

#458 of 1199Rank 458 of 1199 measured models

Gemma 4 12B (Non-reasoning) scores 9.4 on the Artificial Analysis Intelligence Index, ranking 446th of 744 measured models. The closest models on this index are DiffusionGemma 26B A4B (9.5), Qwen3 VL 30B A3B (Reasoning) (9.5), and Gemini 2.0 Flash Thinking Experimental (Jan '25) (9.4). At $0.15 per million tokens (3:1 input/output blended price), it is cheaper than Qwen3 VL 30B A3B (Reasoning) ($0.50). With an output speed of 111 t/s, it is about as fast as Qwen3 VL 30B A3B (Reasoning) (114 t/s).