OpenAI CEO Sam Altman said on Thursday that the company's next large-scale ai model may not be released this year because the company is "prioritizing delivery" of existing models focused on reasoning and difficult problems.

Altman wrote on Reddit AMA: "All these models have become quite complex, and we can't release as much as we'd like at the same time. He said that the company faces "constraints and difficult decisions" when allocating computing resources for "many great ideas."

A Reddit user asked Altman about the release date of GPT-5 or an equivalent version. OpenAI released GPT-4 in March 2023.

Altman responded that OpenAI will release "some very good versions later this year," but "we won't call it GPT-5."

OpenAI faces pressure to continue launching faster, more efficient updates as its competitors, from internet giant Google to well-funded startups like Anthropic, strengthen their ai models. Earlier this month, OpenAI completed a latest funding round with a valuation of $157 billion. The company expects to generate $3.7 billion in revenue this year, with a loss of approximately $5 billion.

Reddit users asked OpenAI executives about the release dates for other new models and features, including the Sora video generation model, new versions of the image generation model DALL-E, ChatGPT's "camera mode," and the visual features of the Advanced Voice Mode (AVM) in ChatGPT.

One questioner asked whether the delay of Sora is due to the "computational volume/time required for reasoning, or due to security issues." OpenAI's product chief Kevin Weil wrote, "Models need to be perfected, ensuring safety/simulation/correctness in other aspects, and need to scale computing!"