o1 vs Qwen3.5 4B
18 SHARED BENCHMARKSAcross 18 shared benchmarks, o1 scores higher on 13 and Qwen3.5 4B on 5. The widest gap is OmniScience Accuracy, where o1 scores 34.7 against 15.1. Qwen3.5 4B is the cheaper of the two on tracked API pricing ($0.03 against $15.00 per million input tokens).
OPENAIVSALIBABA18 SHARED13–5 HEAD-TO-HEAD
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, direct), otherwise the lowest tracked offer.