OpenAI drops a lightweight Deep Research for everyone

OpenAI just announced a lightweight version of its deep research feature, bringing the agentic research capability to a much broader audience.

According to OpenAI's post on X, the new lightweight version is designed to boost current rate limits for Plus, Team, and Pro users. Even better for non-paying users, the company is also rolling this lighter version out to Free accounts.

More research, less compute

Original deep research tasks consume massive compute by chaining together multiple web searches, reading lengthy sources, and synthesizing comprehensive reports, which previously forced OpenAI to keep usage limits tight.

While OpenAI hasn't disclosed technical specifics on what was trimmed under the hood, introducing a lightweight tier lets power users run far more research queries without hitting immediate walls, all while opening the door for casual and free users.

Why it matters

We think this is a smart tiering move by OpenAI.

Full-scale research agents are notoriously expensive to run. By creating a lightweight option, OpenAI balances infrastructure costs against heavy user demand, preventing paid subscribers from running out of query limits during iterative research sessions.

For regular and free users, it removes the paywall on multi-step agentic workflows. That said, it remains to be seen how much nuance, depth, or cross-checking is sacrificed in this streamlined version compared to the heavy-duty original.