OpenAI launched GPT-6 Astra today. Officially it is supposed to be good at science, along with computer use and software.
That is useful. It is not the lab's week of record.
ResearcherFlow will use whatever models are good at scientific research. Today that is Claude Sonnet through OpenRouter. We may add GPT-6 Astra later. A better science engine still has to read the literature you imported and the log you kept.
A model that can reason about a paper still does not write three lines when the run ends. It does not know which student is blocked unless someone logged it. It does not assemble Monday from a workspace that does not exist.
The shareable unit is still a paper card: contribution, methods, findings, limitations. Flow can still search the set, draft a synthesis, and stop until you approve it. A better engine underneath does not change that gate.
Subscribe on this page if you want the next one.
Jon Marrs
ResearcherFlow
https://researcherflow.com