VECTOR WIREAI INTELLIGENCE
UTC
Refresh Models Deals Regulatory Sources

DeepSeek-V4-Flash vs Step 5 Preview

11 SHARED BENCHMARKS

Across 11 shared benchmarks, DeepSeek-V4-Flash scores higher on 1 and Step 5 Preview on 10. The widest gap is OmniScience Non-Hallucination, where Step 5 Preview scores 57 against 15.6. DeepSeek-V4-Flash is the cheaper of the two on tracked API pricing ($0.44 against $1.00 per million input tokens).

DEEPSEEKVSSTEPFUN11 SHARED110 HEAD-TO-HEADNEWEST SCORE ADDED
AA-LCR79.788.3
critpt16.620.9
gdpval52.953.3
HLE38.646.5
MMMU-Pro74.876.4
scicode50.358.9
terminalbenchV4012.133.3

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, direct), otherwise the lowest tracked offer.

Compare DeepSeek-V4-Flash withALL PAIRINGS →