Which is better, GPT-4o or Qwen2.5 Omni 7B?

Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, GPT-4o or Qwen2.5 Omni 7B?: Across 15 shared benchmarks, GPT-4o scores higher on 12 and Qwen2.5 Omni 7B on 3. The widest gap is GPQA Diamond, where GPT-4o scores 70.1 against 30.8. Qwen2.5 Omni 7B is the cheaper of the two on tracked API pricing ($0.10 against $5.00 per million input tokens).

GPT-4o vs Qwen2.5 Omni 7B

Across 15 shared benchmarks, GPT-4o scores higher on 12 and Qwen2.5 Omni 7B on 3. The widest gap is GPQA Diamond, where GPT-4o scores 70.1 against 30.8. Qwen2.5 Omni 7B is the cheaper of the two on tracked API pricing ($0.10 against $5.00 per million input tokens).

OpenAIvsAlibaba15 shared benchmarks12–3 head-to-head

Benchmark	GPT-4o	Qwen2.5 Omni 7B
AI2D	94.2	83.2
chartqa	88.1	85.3
docvqa	92.8	95.2
EgoSchema	72.2	68.6
GPQA Diamond	70.1	30.8
GSM8K	95.6	88.7
humaneval	90.6	78.7
mathvision	30.4	25
MathVista	63.8	67.9
mmlu_redux	88	71
MMLU-Pro	74.7	47
MMMU	72.2	59.2
MMMU-Pro	59.9	36.6
MMStar	63.9	64
RealWorldQA	75.4	70.3

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