Opera Neon Introduces ODRA: A Powerful AI-Driven Deep Research Agent

Opera expands Neon's AI toolkit with a deep research agent - 9to5Mac

Opera's AI-powered browser can now split queries into multiple tasks to quickly provide a more comprehensive result. Here's how that works. A few weeks ago, Opera Neon launched with three browsing agents: Do, Make, and Chat, which can act on the user's behalf to browse the web, gather information, and deliver anything from a quick summary to an interactive widget. Now, Neon is getting a fourth agent, called Opera Deep Research Agent, or ODRA, for more comprehensive research tasks. Here's how Opera presents the new agent: Well, the main takeaway is that we managed to parallelize the operation on the server side, which means we're dividing the problem (your research query) into smaller ones and running separate "researchers" on them - it's like we're applying brains instead of muscle into the engine to come up with a more efficient agent. This basically means that our agent is not going about your research request in a "one by one, and step by step" process, but rather applying a "division of labor" logic to work on various things at the same time (parallelizing). Think about it as the main difference between the way in which a CPU and a GPU work: one is great for performing one single operation in rapid succession, and the other is great at performing many smaller operations simultaneously. Interestingly, once the subtasks are completed, a separate AI, called "supervisor" analyzes the material and either decides that the information is sufficient to what the user asks, or instructs the agent to gather further information to provide and even more comprehensive result. Opera says that each deep research session "typically it takes from 5 to 20 minutes," and that ODRA performed particularly well in the DeepResearch benchmark, second to Google's Gemini 2.5 Pro Deep Research model. For Neon users, ODRA will be integrated into the Omnibus, alongside Made, Chat, and Do. The company offered a sneak peek of what to expect form a prompt such as "Research and analyze the latest advancements and cutting-edge theories within the field of game design. Specifically include recent developments, research, and practical design applications related to established frameworks like MDA (Mechanics-Dynamics-Aesthetics)" on its blog. Opera Neon requires a $19.99 monthly subscription. You can visit Opera Neon's website to join the waitlist. Are you interested in agentic browsing? Let us know in the comments.

Opera's $19.99-a-month AI browser now has a deep research agent

Opera has expanded its AI-powered browser, Opera Neon, by introducing the Opera Deep Research Agent (ODRA). This new component is designed to conduct comprehensive research by splitting user queries into parallel tasks for more thorough results. The addition of ODRA follows the launch of Opera Neon several weeks ago, which initially included three browsing agents: Do, Make, and Chat. These agents were created to act on behalf of the user to browse the web, gather information, and deliver content ranging from summaries to interactive widgets. ODRA now joins this suite as the fourth specialized agent, focused specifically on in-depth research requests. The core function of ODRA involves a method of parallel processing on the server side. Opera described the process, stating, "we're dividing the problem (your research query) into smaller ones and running separate 'researchers' on them - it's like we're applying brains instead of muscle into the engine to come up with a more efficient agent." The company elaborated that this approach avoids a sequential, step-by-step process, instead applying a "division of labor" logic. This method is compared to the operational difference between a CPU, which performs single operations in rapid succession, and a GPU, which simultaneously performs many smaller operations. Once the parallel subtasks are finished, a distinct AI component, referred to as a "supervisor," evaluates the collected material. This supervisor then makes a determination on whether the information is sufficient to fulfill the user's original request. If the results are deemed incomplete, the supervisor instructs the agent to gather additional information to generate a more comprehensive final output. According to Opera, a standard deep research session using ODRA typically requires between 5 and 20 minutes to complete. The agent's performance was measured in the DeepResearch benchmark, where it achieved the second-highest ranking. The only model to score higher was Google's Gemini 2.5 Pro Deep Research model. For individuals using Opera Neon, ODRA will be integrated into the browser's Omnibus interface, functioning alongside the existing Make, Chat, and Do agents. The company provided a sample of a complex prompt the agent can handle, which asked it to "Research and analyze the latest advancements and cutting-edge theories within the field of game design. Specifically include recent developments, research, and practical design applications related to established frameworks like MDA (Mechanics-Dynamics-Aesthetics)." Access to Opera Neon and its suite of AI agents is available through a monthly subscription priced at $19.99. Potential users can register for a waitlist via the official Opera Neon website.