OpenAI is preparing to launch GPT-5.4, the next version of its large language model (LLM). While no exact date of launch is out yet, the ChatGPT-maker says that it is coming “sooner than you think.” While you wait, you do get the option of GPT-5.3 Instant, which is seemingly designed to address a few well documented shortcomings of previous GPT-5 iterations specifically around its tone and accuracy. Meanwhile, online reports suggest that GPT-5.4 will offer a large context window and extreme reasoning capabilities.
According to The Information, GPT-5.4 will unlock a larger context window for users. More precisely, it is reported that it will come with one million tokens (some reports suggest up to 2 million tokens could be in the offing as well) which is a big jump from the 400,000 tokens available in GPT-5.2. In other words, it will be able to ingest more data in a single prompt something that should expedite coding tasks as well as deeper research.
The new model in fact appears to be tailor-made for research with the report hinting at an “extreme” reasoning mode that would allow ChatGPT to answer more complex queries, though it is likely to come at the expense of time and compute. More details are awaited.
In the meanwhile, OpenAI has released GPT-5.3 Instant. The announcement comes at a time when the ChatGPT-maker is under fire for signing a deal with the US Military even as rival Anthropic decided to walk out of it over moral and ethical grounds. The agreement makes OpenAI the default choice for AI deployments in Pentagon operations though the company has briefly paused deployment to NSA and other defense agencies so it could make few changes before going ahead with the contract. The bone of contention is the use of AI for mass surveillance and development of autonomous weapons.
Anthropic’s loss didn’t result in any short-term gains for OpenAI. Ever since Sam Altman swooped in to sign the same contract that Anthropic declined, hoping it would de-escalate the tensions between the US Government and the AI industry, ChatGPT users have been moving out in swarms to Claude and other AI chatbots resulting in almost 300 per cent uninstalls over the weekend (even as Claude became the number one app on the Apple US App Store).
Sam Altman has acknowledged that he might have handled the situation in a rushed and sloppy manner, and the agreement signing might have looked opportunistic to outsiders, the comments have done little to pacify users and even some OpenAI employees who have either quit or shown support for Anthropic protesting the use of AI in military surveillance.
OpenAI would be hoping GPT-5.4 and GPT-5.3 Instant (which is billed to be less cringe and more useful) in the interim will persuade existing users to stay and those who left to come back.
