ChatGPT will soon become faster and more conversational because OpenAI updates his flagship GPT-5 model to GPT-5.1.
OpenAI announced two updates to the GPT-5 series: GPT-5.1 Instant and GPT-5.1 Thinking. Both models are now available on ChatGPT.
GPT-5.1 Instant, essentially the default and most widely used model, is now “warmer, smarter, and better at handling instructions,” according to OpenAI. Meanwhile, GPT-5.1 Thinking is an advanced reasoning model that responds faster to plain tasks and more persistently to intricate ones.
“We clearly heard from users that great AI should not only be smart, but also fun to talk to,” OpenAI said in blog entry. “GPT-5.1 significantly improves both intelligence and communication style.”
The company added that both models enable users to “shape the tone of ChatGPT”, allowing users to control how the chat platform responds depending on the conversation they are having.
Both models have been made available to ChatGPT Pro, Plus, Go and Business users, as well as in the free tier. Those on the Enterprise and Edu plans will receive seven days early access to the models before GPT-5.1 becomes the default model. OpenAI said the models could also be accessed via API, both with customized reasoning.
OpenAI has noted that it will update GPT-5 Pro to version 5.1 soon.
Immediate and thinking models
TTag 5.1 reflects improvements to the base model, and OpenAI considers these parts of the GPT-5 family trained on the same stack and data as the reasoning models. The biggest difference between version 5.1 and version 5 is its more natural and conversational tone, OpenAI general manager of applications Fidji Simo said on Substack post.
“Based on early testing, it often surprises people with how fun it is while remaining clean and useful,” OpenAI said in its post.
Instant can exploit adaptive reasoning to aid him decide when he needs to think about his answers, especially when it comes to more complicated questions. OpenAI noted that it has improved following model instructions so that it still responds quickly, but at the same time responds directly to user queries.
The latest models, e.g BaiduERNIE-4.5-VL-28B-A3B-Thinking, I have was better than GPT-5 in benchmarks such as following instructions.
GPT-5.1 Thinking can independently determine how much reasoning power it should devote to a prompt. It adapts to the type and complexity of the request, so a fuller, intricate question will take longer to answer than a plain summary request.
OpenAI said that evaluations showed that GPT-5.1 Thinking spends less time and therefore uses fewer tokens for plain tasks compared to GPT-5, outperforming the base model in terms of responsiveness.
One thing enterprises should note is that GPT-5.1 thinking equates to “less jargon and fewer undefined terms.” OpenAI claims that removing jargon answers makes Thinking more accessible when it comes to explaining technical concepts.
More personalization
Another gigantic update to ChatGPT is increased customization. This allows users to switch conversations between a genial and reliable chat platform.
ChatGPT already allows users to select pre-built model tone options, but the novel update expands these options “to better reflect the most popular ways to use ChatGPT.”
ABOUTavailable options include “default”, “friendly” (formerly “listener”), “efficient” (formerly “robot”), “professional”, “honest”, and “quirky”. The other two personalities, “cynical” and “nerdy”, remain unchanged.
“We think many people will find that GPT-5.1 better combines IQ and EQ, but one default clearly cannot meet everyone’s needs,” Simo said. “That’s why we’re also making it easy to customize ChatGPT with a range of presets to choose from. The model has the same capabilities whether you choose the default option or one of these options, but the style of its response will vary – more formal or familiar, more playful or direct, more or less jargon or slang. Of course, the eight personalities still don’t cover the full range of human diversity, but we know from our research that many people prefer simple, guided control over too many settings or open-ended options.”
Users can also adjust how much emojis ChatGPT uses. OpenAI offers granular response control and is experimenting with the ability to make models more concise, hot and scannable.
Saving the deployment
OpenAI’s GPT-5 implementation was…less than perfect. While company executives, including CEO Sam Altman, touted the novel model’s capabilities, the decision to initially phase out older and beloved models from ChatGPT was met with dissatisfaction. Worse still, many early adopters found that GPT-5 performed no better than older options in areas such as math, science, and writing.
This prompted Altman to backtrack on some of his statements regarding model deletion, blaming the GPT-5 router for the performance issues. A router that automatically routes queries to the most appropriate models will not be a thing of the past, as GPT-5.1 Auto will direct prompts to the model type that best answers the queries.
OpenAI notes that the GPT-5 Instant, Thinking and Pro models are still available in the ChatGPT model drop-down list, although paid subscribers only have three months to compare these older versions with the 5.1 update. However, models such as GPT-4o will not be affected by the GPT-5 expiration period.
“In the future, as we roll out new ChatGPT models, our approach is to provide ample space for people to evaluate changes and share feedback, allowing us to continue to innovate on our pioneering models while transitioning seamlessly,” the company said. “Sunset times will be communicated clearly and well in advance.”