Awesome-Deep-Research

Report Abuse

Basic Information

This repository is a curated, up-to-date collection of resources about Agentic Deep Research aimed at researchers, developers, and enthusiasts who want a single reference for the rapidly evolving area of autonomous research agents. It aggregates industry product summaries, open-source implementations, recent and influential research papers, evaluation benchmarks, and contribution guidance. The README organizes content into sections such as industry-leading products, open-source projects, latest papers, and benchmarks, and highlights a position paper on the shift from web search to agentic research. The collection emphasizes systems that combine language models, search, retrieval and reasoning and includes entries for multi-agent and single-agent research platforms. The repo is maintained as an 'awesome' list to help users discover implementations, papers and evaluation suites relevant to agentic deep research.

Links

Categorization

App Details

Features
Curated links to industry products and research platforms, an extensive table of recent research papers with metadata such as base models, optimization and evaluation datasets, and a long list of open-source implementations and frameworks. It catalogs benchmarks and practical applications with references to evaluation suites used for agentic search and browsing. The README includes contribution instructions and citation examples for academic use. Visual badges and asset images are used for navigation. Entries often include short descriptions and repository references, and the papers section is organized chronologically with columns for model, training and evaluation details to help readers compare approaches and priorities in the field.
Use Cases
The repository centralizes scattered information about agentic deep research so users can quickly find implementations, state-of-the-art papers, and relevant benchmarks. It helps researchers survey recent trends in search-enhanced reasoning, multi-agent architectures, and retrieval-augmented generation by listing model bases, optimization methods, and evaluation datasets. Developers can discover open-source projects and integration examples for building or experimenting with deep research agents. The benchmarks and application section aid in selecting suitable evaluation suites for experiments. Contribution guidelines and citation snippets make it easier for the community to extend the collection and reference the work in academic outputs.

Please fill the required fields*