Introducing jina-deepsearch-v1, it search, read, reason, search, read, reason, search, read, reason, search, read, reason, search, read, reason, search, read, reason, search, read, reason, search, read, reason, search, read, reason, search, read, reason, ... 🔄 until the best answer is found.

DeepSearch is our first agentic search system with reasoning and planning. Perfect for complex queries. Give it a vibe check here or try it in your fav local chat client such as Chatwise, Cherry Studio, Chatbox. Our API is fully compatible with OpenAI Chat API schema - so simply swap the domain works!

btw, DeepSearch is also open source, you can find the code here: Feel free to contribute or simply let us know what you find interesting!

Comments pointed out that there's a factual mistake in the first video, where the system failed to understand that "full deepseek" means the 671B model. So here's the 2nd try; watch till the end:

Why search some of the Internet, when you can search all of the Internet? All the Internet was created to be extremely fast, easy to use, secure and have the most relevant search results. You can search the whole Internet, with All the Internet.

bold move showcasing your product outputting complete misinformation with no distinction between r1 distilled models and full r1

Isn’t the correct answer around 1.5TB of VRAM for the full model? 🤔

Very nice! I have Iterative search and reasoning over Google Workspace(gmail, drive, attachments, contacts, events etc) with DeepSeek R1. Iterative RAG is quite powerful

Nice!

Great. Thank you

seesh 🔥🔥🔥

Thanks for sharing