OpenAI’s specialized GPT-6 research model for multistep life-sciences work, combining biological reasoning with scientific tool use and offered through controlled access to approved Research customers.
API model ID: `gpt-6-rosalind`. As of October 30, 2026, OpenAI is rolling it out to approved Research customers in Work, Codex, and the API, but not Chat. Enterprise workspaces begin with a two-week Admin Preview that is off by default; organization-level Research approval and authorized-user eligibility are still required.
Only published specifications are shown.
— OpenAI began offering GPT-6 Rosalind Research to approved Research customers in Work, Codex, and the API, but not Chat; Enterprise workspaces start with a two-week Admin Preview that is off by default. API ID: `gpt-6-rosalind`. It uses the same weights as GPT-6 Astra, with no additional training data or post-training; its main distinction is life-sciences safeguards and controlled access. ↗
Its distinct research-model name and API ID, together with specialized deployment safeguards for life-sciences work, merit a separate record. OpenAI explicitly says it uses the same weights as GPT-6 Astra with no additional training data or post-training. This record therefore represents a research-access and safeguards variant of Astra’s weights, not a new checkpoint, and does not copy specifications that were not verified for this model.
OpenAI began offering GPT-6 Rosalind Research to approved Research customers in Work, Codex, and the API, but not Chat; Enterprise workspaces start with a two-week Admin Preview that is off by default. API ID: `gpt-6-rosalind`. It uses the same weights as GPT-6 Astra, with no additional training data or post-training; its main distinction is life-sciences safeguards and controlled access.
View sourceSelect a card to view the related model.
Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.