Across 32 shared benchmarks, DeepSeek-V4-Pro scores higher on 29 and Kimi K2 Instruct on 3. The widest gap is HLE, where DeepSeek-V4-Pro scores 48.2 against 7.9. Kimi K2 Instruct is the cheaper of the two on tracked API pricing ($0.60 against $1.32 per million input tokens).
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.