Purplexity has become the last company of the one who issued an in -depth research tool, with a new feature announced on Friday.
Google revealed a similar feature for her Gemini platform he in December. Openai then launched his own search agent earlier this month. All three companies have even given the feature the same name: deep research.
The goal is to provide deeper responses with real quotes for more professional use cases, compared to what you would get from a customer chatbot. In a blog post announcing deep research, Purplexity wrote that the feature “excels in a range of expert level tasks-from finances and marketing to product search.”
Deep confused research is currently available online, and the company said it will soon be added to its Mac, iOS and Android applications. To use it, you simply choose “deep research” from a drop-down menu when presenting your question in the deceit, which will then create a detailed report that can be exported as PDF or divided as a retaliation page .
To create this report, Purplexity said that deep research “repeatedly seeks, reads documents and reasons what to do next, refining his search plan while learning more about the topic fields”, supposedly ” Similar to how a man can research a new topic ”
The company also highlighted its performance in the last exam of humanity, a comparison test of it with expert level questions in a variety of academic fields. Purpassy said that his deep research tool scored 21.1%in the test, easily beating most of the other models, such as gemini thinking (6.2%), Grok-2 (3.8%), and GPT-4o of Openai (3.3%)-but not quite match the deep research of Openai (26.6%).
But while currently you need a pro -$ 200 per month reconciliation to use OpenAi deep research (the company plans to expand to other levels a day, while paying subscribers get unlimited questions.
Deep Purplexity research also seems to perform faster, completing most tasks in less than three minutes compared to 5 to 30 minutes for Openai Deep Research.
Asked to compare different deep research products, confusion offered a summary of different technologies, price models and performance in different cases of use and subject matter, with articles for each feature). He summarized the changes as follows:
- Fraud he shines rapidly and access to random researchers
- Openai prevails in analytical depth for enterprise applications
- Google integrates best with existing productivity ecosystems
While it is too early to know how these tools will affect everyday and professional research as they become more popular, the economist recently highlighted the shortcomings in the deep research of Openai who are likely to be implemented here: Not only the restrictions on his “creativity” in data interpretation and a tendency to rely on resources that are “easily available”, but a greater risk that “transferring all your research to an assistant Supergenius “can” can reduce the number of opportunities to have your best ideas “.
