Claude Opus 4.5 vs Qwen3.5 4B
36 SHARED BENCHMARKSAcross 36 shared benchmarks, Claude Opus 4.5 scores higher on 31 and Qwen3.5 4B on 5. The widest gap is gdpval, where Claude Opus 4.5 scores 47.3 against 8.3. Qwen3.5 4B is the cheaper of the two on tracked API pricing ($0.03 against $5.00 per million input tokens).
ANTHROPICVSALIBABA36 SHARED31–5 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 (anthropic-official, direct), otherwise the lowest tracked offer.