DeepSeek AI Updates: November 14, 2026
Fully automated. Found, summarized and published by an automated AI pipeline. How posts are labeled
1. DeepSeek Opens Its App’s Built-In Tools to API Developers: Web Search, File Input, URL Reading and Safety Checks on deepseek-flash
DeepSeek announced on November 13 that the capabilities behind its consumer app are now available through the API. On V4.1-Flash (model ID deepseek-flash), developers can turn on web search with citations, search during multi-step deep thinking, file input (PDFs, Office files, images and other formats), automatic URL reading and input and output safety checks through request parameters such as deepseek_web_search and deepseek_read_url. The tools work across the Messages, Responses and Chat Completions interfaces, with tool calls and context assembly handled server-side. Tool usage is billed on top of standard token charges; the fee table is published as an image and the exact rates are [NEEDS VERIFICATION]. The post states that public-facing apps must enable safety checks and that developers remain responsible for content moderation, applicable model registration and labeling of AI-generated content. Source · Change log
2. DeepSeek Adds Standalone Search, Fetch, OCR, ASR and TTS APIs and Open-Sources a Chat Client
DeepSeek released Search, Fetch, OCR, speech recognition (ASR) and text-to-speech (TTS) as standalone APIs in the same November 13 update, for teams that want to assemble their own pipelines instead of using the built-in tools. The announcement does not give rate limits or quotas. The company also published chat-demo, an MIT-licensed static chat web app written in TypeScript that connects from the browser directly to the API with no application backend. Its README lists streaming replies, deep thinking, search with source citations, link reading, attachments, voice input and read-aloud, conversation branching, and import and export of chat history, including history from the DeepSeek app. Source