DeepSeek to Launch V4.1 Flash AI Model, Surpassing V4 Pro in Performance and Cost

DeepSeek Announces V4.1 Flash Release

DeepSeek is scheduled to release its V4.1 Flash model around September 10, 2026 (Beijing Time). This new model has undergone extensive internal and external testing.

The V4.1 Flash model has demonstrated superior performance across key metrics, including overall performance, cost efficiency, processing speed, and task completion time, compared to the existing V4 Pro model.

Automatic Upgrade for Pro Users

Following the official launch of V4.1 Flash and prior to the release of V4.1 Pro, all requests directed to the V4 Pro model will be automatically rerouted to V4.1 Flash. These requests will also be billed at the V4.1 Flash pricing.

This change aims to provide users with the benefits of the newer, more capable model at a reduced cost without requiring manual migration.

New Pricing Structure for Flash Series

DeepSeek will implement new pricing for its Flash series, effective from 12:00 Beijing Time on September 10, 2026. During off-peak hours, the unit price for input cache hits will be $0.003, input cache misses $0.15, and output $0.6.

Peak-hour usage will incur charges that are double the off-peak rates. Users are advised to consider these new rates when planning their model usage.