OpenAI has announced multiple product updates this week, with insiders saying "fresh out of the oven and immediately live," and GPT-5.5 may follow shortly.

ME News message: On April 23 (UTC+8), according to Beating monitoring, OpenAI entered a high-density release cadence this week, rolling out multiple product updates in just a few days: Workspace Agents (GPTs upgraded to team-shared agents), ChatGPT for Clinicians (a physician-focused version, free), ChatGPT Images 2.0 (2K resolution, dual-mode), the open-source local de-identification model Privacy Filter, Codex screen-observation memory Chronicle, and the open-source Harmony conversation visualization tool Euphony.

Multiple signals point to the impending launch of GPT-5.5. The GPT-5.5 name previously briefly appeared at the top of the Codex model selector. Brockman also teased on this week’s podcast that “there will be a new model to come in the writing dimension.” Observers widely speculate that OpenAI’s next-generation frontier model under the internal codename “Spud” is GPT-5.5, and that pre-training has reportedly been completed by the end of March.

Today, on X, Jason Liu from OpenAI’s developer experience team and research scientist roon responded to external doubts about the release pace. Some users criticized OpenAI for “squeezing out features” ahead of the “Spud” launch; roon replied that the model and features will be released as soon as they’re ready—“it looks like we’re squeezing them out because they’re actually being completed one after another.” Jason Liu, speaking from an insider’s perspective, confirmed: before joining, he also thought the company would hold models back until competitors arrived or until the features were fully ready—“that’s not the case; everything goes live as soon as it’s out.”

(Source: BlockBeats)