DeepSeek Updates V4 Pro API to Support Responses API
DeepSeek's API documentation now includes the deepseek-v4-pro model, featuring support for the Responses API and tool calls. The model boasts a context length of 1 million tokens.
DeepSeek has updated its large language model, DeepSeek V4 Pro, by adding support for the Responses API. The new model version, identified as DeepSeek-V4-Pro-0813, is now available for API users.
The expanded context window allows for the processing of up to 1 million tokens, with a maximum output limit of 384,000 tokens. The model also supports tool calls, enabling integration with other systems.
Pricing for the service is set at $0.003625 per million input tokens for cache hits and $0.435 per million for cache misses. Output tokens are priced at $0.87 per million.
This update enhances the model's capacity and functionality, addressing the growing demand for processing large datasets within AI applications.