jina-ai / jina-ai/node-DeepResearch
thinking model
Open
- Dominant language
- TypeScript
- Stars
- 5.2k
- Forks
- 461
- PR merge metrics
- No merged PRs in 30d
Description
Check out this implementation: https://github.com/dzhng/deep-research/
It turns out that you dont really need structured output from the LLM (meaning we can use thinking models) for this kind of project. We just have to prompt them to output in JSON format. Does it work 100% of the time? No. But is it five times smarter? Yes.
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.