.webp)
GPT-5 is coming – and rumors dictate its release date will be sooner rather than later.
In his podcast interview with Bill Gates, OpenAI CEO Sam Altman confirmed in January 2024 that GPT-5 was under development. Reports suggest that we may see it sooner than we think: two anonymous sources close to OpenAI suggested GPT-5 would be released mid-2024, likely this summer.
What is GPT-5?
While GPT-4 seems revolutionary to us now, Altman believes the world has only scratched the surface of AI. At the World Government Summit in January 2024, Altman compared the current models from OpenAI to the early days of cell phones:
If one thing’s for certain, it’s that the next generation of GPT models is unimaginable to us right now. While it will take time to get from the flip phone version of GPT to the iPhone version, we’ll be one step closer by the end of the year.
GPT-5 will be the most-anticipated LLM release to date. The AI model will have more speed and heightened natural language processing capabilities. So what does it hold in store?
"It's really good, like materially better," said one CEO with advanced GPT-5 access.
Insights from OpenAI
As the GPT-5 launch nears, OpenAI execs and insiders have increased their comments on the next-gen model. Here's what the company has stated recently:
- CEO Sam Altman claimed that GPT-5 will be able to process emails and calendar details, and that it will be more customizable
- CTO Mira Murati explained in a Dartmouth Engineering interview that GPT-3 had the intelligence of a toddler, GPT-4 was more similar to a smart high-schooler, and that the next generation will look to have PhD-level intelligence (in certain tasks)
- Microsoft AI CEO Mustafa Suleyman shared that it won't be until GPT-6 in two years' time that the models will be able to 'take action' in novel environments
What’s the difference between GPT-4 and GPT-5?
Just as GPT-4 was a sizable increase from its predecessor, there’s no doubt the next version will do the same. While no updates have been confirmed by OpenAI – and likely won’t be before its surprise release – the following predictions of what to expect from GPT-5 have been sourced from insider interviews, previous GPT updates, and AI trends from competitor LLMs.
It’s smarter than GPT-4
The biggest difference? According to Altman: “It’s going to be smarter.”
At the World Government Summit, Altman underscored that the biggest leap from GPT-4 to GPT-5 was deceptively simple: it’s more intelligent. “What makes these models so magical is that they’re general,” he explained. “It’s going to be smarter, so it’s going to be better at everything across the board.”
While there are plenty of improvements expected – new features, faster speeds, and multimodalism, according to Altman’s interview – a more intelligent model will enhance all existing features of current LLMs.
Increased reliability
Altman confirmed that reliability will be a core focus of GPT’s (Generative Pre-trained Transformer) evolution over the next two years. “If you ask GPT-4 most questions 10,000 times, one of those 10,000 is probably pretty good, but it doesn’t always know which one, and you’d like to get the best response of 10,000 each time, and so that increase in reliability will be important.”
Reliability has long been a sticking point for GPT-4 users, with GPT-4 Turbo developed partially to make necessary updates to the model’s output consistency and accuracy. Turbo saw an increase in model stability and fewer AI hallucinations.
So if well-founded user complaints are heard, the new model will almost certainly be trained to be more reliable than GPT-4. In fact, an anonymous OpenAI insider confirmed that the team is hoping for increased reliability in the next iteration. We’re hoping so, too.
Enhanced reasoning abilities
At the center of its general intelligence is GPT-5’s more advanced ability to reason. “Maybe the most important areas of progress will be around reasoning ability,” Altman shared with Gates. “Right now, GPT-4 can reason in only extremely limited ways.”
There’s no shortage of users posting their GPT-4 fails on Reddit and Medium, from group roasts of its problem-solving, to formal explanations of its limited reasoning capabilities. It’s easy to understand why: reasoning is notoriously difficult. Any materially significant improvements would lead to huge strides for the AI model’s performance.
Improved reasoning would mean GPT-5 would be better at understanding context, making inferences, and problem-solving than GPT-4. Combined with a larger knowledge base, it would mean GPT-5 is better able to understand user intent and follow up with more relevant information.
And if we’re lucky, GPT-5 will be the model that finally figures out how to answer riddles, propelling it far beyond GPT-4.
More multimodal
Multimodality has been central to the past few iterations of GPT. OpenAI shows no signs of slowing it down.
OpenAI introduced GPT-4o in May 2024, bringing with it increased text, voice, and vision skills. A far stone’s throw from GPT-4 Turbo, it’s able to engage in natural conversations, analyze image inputs, describe visuals, and process complex audio.
Changes in multimodality create huge shifts in how we engage with GPT. Natural conversation flow – when the model can accurately interpret tonal changes and follow human-like speech patterns, like GPT-4o – is a giant leap in AI natural language processing.
And it’s not just heightened voice and text. OpenAI hasn’t been shy to tease their upcoming text-to-video model Sora. The AI model was developed to imitate complex camera motions and create detailed characters and scenery in clips up to 60 seconds.
If their history of multimodality isn’t enough, take it from the OpenAI CEO. Altman confirmed to Gates that video processing, along with reasoning, is a top priority for future GPT models.
Multimodality is one of the biggest buzzwords in the future of AI models, and for good reason. Despite GPT-4o’s emphasis on widening its multimodal capabilities, it’d be no surprise to see even more voice, image, or video features with the release of the new model.
Increased parameter size
Each GPT update has increased the parameter size, and the next-generation GPT-5 will likely be no exception. In a transformer like GPT, parameters include the weights and biases of the neural network layers, like the attention mechanisms, feedforward layers, and embedding matrices. The size of these parameters directly influences its capacity to learn from input data.
While OpenAI keeps under wraps their exact parameter size, estimates place it at around 1.5 trillion parameters – a stark increase from GPT-3’s 175 billion (and an unimaginable leap from GPT-2’s 1.5 billion).
AI expert Alan Thompson, an integrated AI advisor to Google and Microsoft, expects a parameter count of 2-5 trillion., which would greatly the depth of tasks it can accomplish for developers. His analysis is based on the doubling of both computing power and training time – a significant increase in testing timeline from GPT-4.
Bigger context windows
Context windows represent how many tokens (words or subwords) a model can process at once. A larger context window enables the model to absorb more information from the input text, leading to more accuracy in its answer.
One of the GPT-4 flaws has been its comparatively limited ability to process large amounts of text. For example, GPT-4 Turbo and GPT-4o have a context window of 128,000 tokens. But Google’s Gemini model has a context window of up to 1 million tokens.
Right now, if your only concern is a large language model that can absorb large amounts of information, GPT-4 might not be your top choice. It’s expected that OpenAI will resolve these discrepancies in the new model.
What does Alan Thompson predict? A whopping increase to 40 trillion tokens. A change of this nature would be a notable advancement over the Gemini model, adding the ability to respond to massive datasets input by users. This would be a game-changer for the AI model’s performance, notably for OpenAI enterprise customers and users with heavy data input needs.
Increased customization
GPT-4 is often used as a one-size-fits-all tool. But future iterations will become more personalized. On Gates’ podcast, Altman reiterated that customizability and personalization will be key to future OpenAI models. “People want very different things out of GPT-4: different styles, different sets of assumptions.”
OpenAI has already introduced Custom GPTs, enabling users to personalize a GPT to a specific task, from teaching a board game to helping kids complete their homework. While customization may not be the forefront of the next update, it’s expected to become a major trend going forward.
In the meantime, you can personalize an AI chatbot equipped with the power of GPT-4o for free. It’s what we do best. Get started here.
When is the GPT-5 release date?
Predictions of a release date have been earnestly estimated by users and journalists alike, ranging from the summer of 2024 to early 2026.
But the most recent (and reliable) update came from OpenAI CTO Mira Murati: In her Dartmouth Engineering interview released on June 20, she claimed the 'next generation' of model was around a year and a half away, putting the release closer to the end of 2025. However, it's unclear whether the next-generation model she discussed with the interviewer was GPT-5, or future generations.
The training period is anticipated to take 4-6 months, double OpenAI’s 3-month training time for GPT-4. The new model will likely undergo reinforcement learning, red teaming, and further testing before being released, although it’s a nebulous timeline – OpenAI may have to shift its launch date further if it encounters unpredicted scenarios in testing.
What training data will GPT-5 use?
GPT-5 will almost certainly continue to use available information on the internet as training data.
If there’s been any reckoning for OpenAI on its climb to the top of the industry, it’s the series of lawsuits about the models’ complete training.
GPT models are trained on massive datasets taken from the internet, much of it copyrighted. This unauthorized use of data has led to widespread complaints and legal action: a lawsuit from The New York Times, a lawsuit from a series of U.S. news agencies, and claims that the model’s training process violates the EU's General Data Protection Regulation.
A California judge has already dismissed one of the OpenAI copyright lawsuits filed by a group of writers, including celebrities Sarah Silverman and Ta-Nehisi Coates. There are no suggestions yet that OpenAI and company will be substantially held back by these complaints as it continues testing.
How much does GPT-5 cost?
If OpenAI continues with their standard pricing model, GPT-5 will cost a premium to use. Currently, ChatGPT with GPT-4 is available only to paying users at $20 per month, while ChatGPT with GPT 3.5 is available for free.
As for API pricing, GPT-4 currently costs $30.00 per 1 million input tokens and $60 per 1 million output tokens (these prices double for the 32k version). If the new model is as powerful as predicted, prices are likely to be even higher than previous OpenAI GPT models.
However, the latest OpenAI model is far more cost-accessible. GPT-4o costs only $5 per 1 million input tokens and $15 per 1 million output tokens. While pricing differences aren't a make-or-break matter for enterprise customers, OpenAI is taking an admirable step towards accessibility for individuals and small businesses.
The good news? The launch of GPT-5 will – fingers crossed – bump GPT-4 to become OpenAI’s new free model.
The Future of ChatGPT
The next generation of large language models will revolutionize how we interact with AI in our day-to-day lives. At Bloomberg’s Tech conference, OpenAI COO Brad Lightcap hinted at how the company plans to revolutionize human-computer interaction, taking GPT from an LLM to a model with agent-like capabilities.
“Will there be such a thing as a prompt engineer in 2026?” Lightcap said. “You don’t prompt engineer your friend.”
A more capable and personalized model with more multimodal capabilities promises just what Altman and OpenAI expect: the unimaginable. The anticipated GPT-5 will be one step closer.
The Power of GPT, Customized
What if your AI chatbot automatically synchronized with every GPT update?
Botpress has provided customizable AI chatbot solutions since 2017, providing developers with the tools they need to easily build chatbots with the power of the latest LLMs. Botpress chatbots can be trained on custom knowledge sources – like your website or product catalog – and seamlessly integrate with business systems.
The only platform that ranges from no code set-up to endless customizability and extendability, Botpress allows you to automatically get the power of the latest GPT version on your chatbot – no effort required.
Start building today. It’s free.
Table of Contents
Stay up to date with the latest on AI chatbots
Share this on:
.svg)
