Which is better, Gemma 3 12B or Qwen2.5 Omni 7B?

Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, Gemma 3 12B or Qwen2.5 Omni 7B?: Across 13 shared benchmarks, Gemma 3 12B scores higher on 9 and Qwen2.5 Omni 7B on 4. The widest gap is GPQA Diamond, where Gemma 3 12B scores 40.9 against 30.8. Gemma 3 12B is the cheaper of the two on tracked API pricing ($0.05 against $0.10 per million input tokens).

Gemma 3 12B vs Qwen2.5 Omni 7B

Across 13 shared benchmarks, Gemma 3 12B scores higher on 9 and Qwen2.5 Omni 7B on 4. The widest gap is GPQA Diamond, where Gemma 3 12B scores 40.9 against 30.8. Gemma 3 12B is the cheaper of the two on tracked API pricing ($0.05 against $0.10 per million input tokens).

GooglevsAlibaba13 shared benchmarks9–4 head-to-head

Benchmark	Gemma 3 12B	Qwen2.5 Omni 7B
AI2D	84.2	83.2
chartqa	75.7	85.3
docvqa	87.1	95.2
GPQA Diamond	40.9	30.8
GSM8K	94.4	88.7
humaneval	85.4	78.7
mathvision	32.1	25
MathVista	62.9	67.9
MBPP	0.7	0.7
MM MTBench	6.7	0.1
MMLU-Pro	60.6	47
MMMU	59.6	59.2
MMMU-Pro	37.5	36.6

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (deepinfra, alibaba-official), otherwise the lowest tracked offer.