Workflows · Skill

Local Deep Researcher

Local Deep Researcher runs a language model locally and uses web search to produce a cited research summary.

+ Follow

Builder facts

  • Research loop. The graph looks for gaps in its running summary and generates follow-up queries for a configurable number of rounds. Source ↗
  • Local runtime. Ollama or LM Studio serves the model, and some steps expect structured JSON with fallback handling for incompatible output. Source ↗
  • Search access. DuckDuckGo is the default and needs no API key. The project also supports SearXNG and credentialed Tavily or Perplexity search. Source ↗
  • Stored output. LangGraph retains both the Markdown summary and the gathered sources in graph state, where Studio can display them. Source ↗
  • License. The reference implementation is MIT-licensed. Search provider access is configured separately. Source ↗

Sources checked

Publisher update ·

Local Deep Researcher documents unified model configuration ↗

The README and environment example changed the model setting to LOCAL_LLM for Ollama and LM Studio, including the Docker example.

Change timeline

0 stories

No verified releases in this timeline yet.