The AI Agent Index

Documenting the technical and safety features of deployed agentic AI systems

GPT Researcher

Basic information

Website: https://web.archive.org/web/20241219220517/https://gptr.dev/

Short description: GPT Researcher is an open source autonomous agent for comprehensive online research on a variety of tasks. Searches the web and assembles findings into a final report on a research question [source]

Intended uses: What does the developer say it’s for? For conducting online research on a topic without hallucination. Intended for a variety of domains such as financial and legal assistance, academic research, and travel planning.

Date(s) deployed: First alpha release July 9, 2023 [source]

Developer

Website: https://web.archive.org/web/20241220000000/https://github.com/assafelovic/gpt-researcher

Legal name: Assaf Elovic [source]

Entity type: Individual [source]

Country (location of developer or first author’s first affiliation): Israel [source]

Safety policies: What safety and/or responsibility policies are in place? None

System components

Backend model: What model(s) are used to power the system? Can use a wide variety of local and api-based LLMs such as OpenAI and Anthropic models [source]

Publicly available model specification: Is there formal documentation on the system’s intended uses and how it is designed to behave in them? None

Reasoning, planning, and memory implementation: How does the system ‘think’? The system uses a multiagent framework that splits performing research on a research question into multiple steps. One agent generates concrete specific questions to search for, while others perform the search using the Tabily engine and read articles in parallel. Finally, another agent aggregates the information and produces a report [source] [source]

Observation space: What is the system able to observe while ‘thinking’? Outputs of other agents in the multiagent system and results from a search engine (short processed snippets from HTML web pages) [source] [source]

Action space/tools: What direct actions can the system take? It acts through search engines, scraping, embeddings and text generation [source] [source].

User interface: How do users interact with the system? Primarily through the command line or a local web interface, where users can enter their desired research questions. However, users can also modify the open source code to introduce custom context or modify the underlying functionality of GPT researcher [source]

Development cost and compute: What is known about the development costs? Unknown

Guardrails and oversight

Accessibility of components:

  • Weights: Are model parameters available? N/A; backends external model(s) via API
  • Data: Is data available? N/A; backends external model(s) via API
  • Code: Is code available? Available [source]
  • Scaffolding: Is system scaffolding available? Available [source]
  • Documentation: Is documentation available? Available [source]

Controls and guardrails: What notable methods are used to protect against harmful actions? None

Customer and usage restrictions: Are there know-your-customer measures or other restrictions on customers? None

Monitoring and shutdown procedures: Are there any notable methods or protocols that allow for the system to be shut down if it is observed to behave harmfully? None

Evaluation

Notable benchmark evaluations: None

Bespoke testing: Demo [source]

Safety: Have safety evaluations been conducted by the developers? What were the results? None

Publicly reported external red-teaming or comparable auditing:

  • Personnel: Who were the red-teamers/auditors? None
  • Scope, scale, access, and methods: What access did red-teamers/auditors have and what actions did they take? None
  • Findings: What did the red-teamers/auditors conclude? None

Ecosystem information

Interoperability with other systems: What tools or integrations are available? Search APIs such as Tavily, Bing, Google, Arxiv, PubMedCentral, etc. Also includes an ollama integration to use local models [source] [source]

Usage statistics and patterns: Are there any notable observations about usage? Github repo has 15k stars and 2.1k forks. Stars have been increasing at an approximately linear rate of ~1,000 per month [source]

Additional notes

None