Which is better, Qwen2.5 Omni 7B or Qwen3 VL Thinking (8B)?

Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, Qwen2.5 Omni 7B or Qwen3 VL Thinking (8B)?: Across 12 shared benchmarks, Qwen2.5 Omni 7B scores higher on 1 and Qwen3 VL Thinking (8B) on 11. The widest gap is mathvision, where Qwen3 VL Thinking (8B) scores 62.7 against 25. Tracked API pricing per million tokens: Qwen2.5 Omni 7B $0.10 in / $6.76 out, Qwen3 VL Thinking (8B) $0.18 in / $2.10 out.

Qwen2.5 Omni 7B vs Qwen3 VL Thinking (8B)

Across 12 shared benchmarks, Qwen2.5 Omni 7B scores higher on 1 and Qwen3 VL Thinking (8B) on 11. The widest gap is mathvision, where Qwen3 VL Thinking (8B) scores 62.7 against 25. Tracked API pricing per million tokens: Qwen2.5 Omni 7B $0.10 in / $6.76 out, Qwen3 VL Thinking (8B) $0.18 in / $2.10 out.

AlibabavsAlibaba12 shared benchmarks1–11 head-to-head

Benchmark	Qwen2.5 Omni 7B	Qwen3 VL Thinking (8B)
AI2D	83.2	84.9
GPQA Diamond	30.8	69.9
mathvision	25	62.7
MathVista	67.9	81.4
MM MTBench	0.1	8
mmlu_redux	71	88.8
MMLU-Pro	47	77.3
MMMU	59.2	73.5
MMMU-Pro	36.6	60.4
MMStar	64	75.3
MV-Bench	70.3	69
RealWorldQA	70.3	73.5

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 (alibaba-official, alibaba-official), otherwise the lowest tracked offer.