DeepSeek V4.1 Flash 1017 (Max) Model Analysis: Intelligence, Speed, and Cost Evaluation

DeepSeek V4.1 Flash 1017 (Max) is a high-performing open weights model with an Intelligence Index score of 45, 227.1 output tokens per second, and $0.30 input/$1.20 output pricing, offering strong intelligence and speed with a 1M token context window. It outperforms many competitors in intelligence metrics but is notably verbose, scoring 210M output tokens compared to the median of 140M.

Comments

Read original article