Abstract:
On September 8, DeepSeek announced in the official communication group today that the intermediate version of DeepSeek V4.1 Flash is under internal testing. According to reports, the model adopts a new model structure, with native multi-modal support, stronger capabilities, faster speed, and lower cost. The base_url of the model remains unchanged, and the model name can be called by setting the model name to deepseek-v4.1-flash-expires-on-0910. The current billing is the same as deepseek-v4-flash, with a limit of 20 concurrency per account.
It is understood that the official version of DeepSeek-V4-Flash API was launched for public testing at the end of July, and the Agent capabilities have been greatly enhanced. On August 21, the DeepSeek-V4-Flash-Vision-Exp multi-modal visual understanding model was launched on the DeepSeek API platform. The model supports inputting images in addition to text, allowing the model to describe images, identify text in screenshots, analyze charts, etc. Supported image formats include JPEG, PNG, GIF, and WebP.
Comments