OpenAI’s new Deep Research agent, developed by the company, promises to revolutionize how users gather online data. It autonomously browses the internet, analyzes responses, and delivers comprehensive documents on any topic. The AI company showcased its capabilities by tackling everything from ski purchase recommendations to advanced biology papers. However, this service is not available to everyone. OpenAI limits access to Pro users who pay a monthly fee of $200.
The high price is justified by the compute-intensive and slow nature of the system. It is the first AI system capable of performing a wide range of complex and valuable tasks, as noted by Sam Altman in a tweet.
However, it is not entirely accurate to claim that OpenAI is the only option available. There have been other options in the market for several months. From open-source projects like the AI Scientist to closed-source projects like Gemini’s Deep Research, users have had the ability to leverage AI for complex research for some time. Those who are unwilling to pay the premium or have a bit of creativity can achieve similar or even better results at a fraction of the cost, or even for free.
Here are some top picks for those on a budget:
1. Google Gemini: The 90% Discount Option
Google introduced Deep Research to Gemini Advanced users in December for a monthly fee of $20. Gemini users gain access to Gemini Ultra, Gemini’s Deep Research Agent, 2TB of cloud storage, and the ability to build personal agents or “gems,” similar to OpenAI’s GPTs. Google’s tool offers unlimited research requests, unlike OpenAI’s query limits for Plus and Pro users. Decrypt’s test of the tool showed that it operates similarly to OpenAI’s approach, with users inputting their topic and receiving a research plan. However, Gemini users must adjust the plan before allowing the AI to browse the web. Gemini’s research process favors established scientific publications and reputable news outlets over smaller sites or blogs. The process usually takes a few minutes as the agent goes through dozens of websites according to its plan.
2. You.com: A Research Agent for Speed
You.com may not be the most well-known AI site, but it should not be underestimated. The startup, backed by notable tech investors, is known as a competitor in the AI-powered search engine space. You.com offers an AI research agent that can analyze up to 200 sources per query and employ PhD-level research methodologies. Users simply type their query, and the model delivers results within seconds or a few minutes. You.com’s reports are shorter than its rivals but provide a solid foundation for further research. The platform offers three free daily uses before requiring users to switch to a different model.
3. Deepseek R1: The Free DIY Powerhouse
Deepseek’s R1 is unique as the only reasoning model with built-in internet access. While it requires more user input, it delivers robust results. Users activate Deep Reasoning and provide detailed instructions on what to search for. R1 shows its work and thought process in real-time, allowing researchers to fine-tune their instructions and understand the model’s thinking. The initial output matches You.com’s scope, but R1’s power shines in follow-up queries. It can provide more detailed information and expand reports within the scope of the topic being researched.
In addition to these major players, other tools like Beago AI, Perplexity Pro, and Agent Laboratory have carved out niches in the AI research space.