Archive — CN — technode
DeepSeek V4 Pro API Update Adds Responses API Support
DeepSeek has updated its V4 Pro API to include support for the Responses API, featuring a context length of 1 million tokens and new pricing structures for input and output tokens. This update is noted in its documentation with the version DeepSeek-V4-Pro-0813.
Key points
- DeepSeek V4 Pro API now supports the Responses API.
- The current version is DeepSeek-V4-Pro-0813.
- It has a context length capability of 1 million tokens.
- Max output capacity is 384,000 tokens.
- Pricing is $0.003625/million input tokens for cache hits, $0.435 for cache misses, and $0.87/million output tokens.
Why it matters
- The update aligns with the trend of enhancing AI APIs for increased efficiency and effectiveness in processing large amounts of data.
- By 2025, the industry trend is moving toward APIs that can handle extensive interactions which this update supports.
- The adoption of large language models (LLMs) is prevalent, as companies seek to enhance their applications with AI capabilities.
The update of DeepSeek V4 Pro API to include the Responses API and expanded functionalities enhances its usability, aligning with increasing demands for advanced data processing in AI applications.