DeepSeek releases Family Bucket application API enabling 1:1 replication of its app
DeepSeek opened its Family Bucket application API on November 13, 2026, letting developers enable deep-thinking search, file input, and other app features via parameters in existing APIs. Voice and underlying capabilities (OCR, ASR, TTS) are available as standalone interfaces. The V4.1-Flash model (deepseek-flash) supports these features across Messages, Responses, and Chat Completions endpoints.
China context
- Original name
- 深度求索
- Outside China
- Global API · api-docs.deepseek.com
- Claims
- Company-reported; not yet independently evaluated
- For builders
- Developers outside China can integrate DeepSeek's search, file parsing, and voice capabilities via API parameters, reducing the need for custom tool orchestration.
- For investors
- DeepSeek's expansion of API capabilities may increase its developer adoption and API usage, potentially impacting the competitive landscape for AI infrastructure providers.
Translated from Chinese. Quotes and facts link to the original sources.
DeepSeek announced the official opening of its Family Bucket application API, integrating the interactive features of the DeepSeek App directly into its existing API. By enabling corresponding parameters, developers can access deep-thinking search, file input, and other conversational capabilities, with tool calls and context assembly handled automatically by the server. The update allows developers to replicate the complete DeepSeek App experience in their own applications. The V4.1-Flash model (deepseek-flash) supports web search, deep-thinking search, file input, link reading, and model safety features across Messages, Responses, and Chat Completions interfaces. Additionally, voice and underlying capabilities such as OCR, ASR, and TTS are available as independent interfaces for flexible integration.
The API exposes app-level features as server-side tools: deepseek_web_search for web/image search in both thinking and non-thinking modes, deepseek_read_url for link reading, and file input supporting PDF, Office, and image formats with text extraction or mixed layout parsing. A safety validation parameter provides basic risk control, returning refusal identifiers when risks are detected. The V4.1-Flash model (deepseek-flash) is the first to support these capabilities across Messages, Responses, and Chat Completions endpoints.
Developers outside China can now build applications with DeepSeek's search, file parsing, and voice capabilities without managing tool orchestration, reducing integration cost and time to market for AI-powered apps. This move pressures competing API providers to offer similar bundled app-level features or risk losing developer mindshare.
For developers, the API reduces the complexity of building conversational AI applications by handling tool calls and context assembly server-side. For DeepSeek, it expands the reach of its app features beyond its own app, potentially increasing API usage and revenue.
A key signal will be the number of public apps built on these new capabilities within the next quarter, and whether other Chinese model providers follow with similar app-level API integrations.