open_deep_research issues

Is it possible to use the ArXiv API instead of perplexity?

7

In the "search_api" configuration is it possible to use ArXiv to retrieve scientific papers and use that information instead of a regular web search?

LuOsorio

Improve answer quality by referring to topics at each step

I ran it with the topic "Compare iPhone 16 vs iPhone 16e". ```python # Fast config with DeepSeek-R1-Distill-Llama-70B thread = {"configurable": {"thread_id": str(uuid.uuid4()), "search_api": "tavily", "planner_provider": "openai", "planner_model": "gpt-4o", "writer_provider":...

moritalous

How do i use a local ollama model as my model here?

1

This is great and looks super powerful, but the cost of so many tokens at scale is going to be huge for heavy use. One great way to cut costs...

Travis-Barton

feat: add demo frontend for open researcher

- **feat: add sample frontend for open researcher** - **Add option to give feedback** - **Break words** - **Expose thinking block** - **Make input box smaller**

dqbd

Are there any benchmark results for this open_deep_research?

1

Please share~

chenatu

Is it possible to use the knowledge of the LLM instead of web search to write the report?

5

Sometimes, I am just interested in extracting the knowledge from the LLM, not on web searches. Is this possible? Also, if it is possible, can we create larger reports (more...

ruipcarvalho

feat: Add Exa, arXiv and PubMed search API integration

Web search using the Exa API with support for: - AI-generated content summaries for search results - Custom result limits and content length controls - Domain filtering (include/exclude) - Subpage...

bartolli

Add support for Google Gemini LLM models

1

Add support for Google Gemini LLM models using the langchain-google-genai package. * **Dependencies** - Add `langchain-google-genai` as a dependency in `pyproject.toml`. * **Configuration** - Add `GEMINI` to the `PlannerProvider` and...

santiago-afonso