Across 6 shared benchmarks, Gemini 3 Pro scores higher on 4 and Qwen3 VL 235B on 2. The widest gap is OmniDocBench v1.5 TextEdit, where Qwen3 VL 235B scores 0.1 against 0.1.
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.