Introducing the Open Deep Research app! Generate detailed reports on any topic with open source LLMs. Free & fully open source. We’re releasing everything: evaluation dataset, code, app, and blog.🔥

Try out the Open Deep Research app here!

More resources on how we built open deep research & the code! - Blog post explaining it: - GitHub repo with the code:

Super super cool! may just be me, but I prefer just responding to the follow-ups in chat for deep research Probably one of: 1. Forms feel more sucky in the AI era 2. I like to answer all the follow ups in 1 sentence and then add other context

Awesome feedback, thank you! We're planning on adding this deep research functionality to and for that one, we're planning on doing follow-ups in-chat (both for clarifying questions & follow ups after report generation).

Can you make this site have PWA functionality? Thanks!

Great work and keeping it free ✌️

Open source LLMs democratize access but raise crucial questions about responsible use.

Seems to be wrong on simple questions: "What is the Total Number of Common Stock Shares Repurchased by Netflix in Q4 2024?" Interestingly it fails in exactly the same way Llama 3.3 did in our finance benchmark. Is there a way to change the underlying model?

Its really really good for an open-source Deep Research. Want to try out more and looking for integration of more models. If wondering below is the example that I have tried: Output: Prompt: