DeepSeek Flash 4.1 is already being tested via API and rolling out.

DeepSeek is internally beta testing an intermediate version of DeepSeek V4.1 Flash, which is now available for API testing. This new model architecture offers native multimodal support, enhanced capabilities, increased speed, and reduced costs. Users can access it by setting the model name to deepseek-v4.1-flash-expires-on-0910, keeping the base_url unchanged. Pricing is currently the same as deepseek-v4-flash, with a rate limit of 20 concurrent requests per account.

This is the first time DeepSeek has offered native multimodal support in its Flash series, unlike previous versions that focused solely on text.