Leak previews four GPT-5 model versions ahead of launch
OpenAI's GPT-5 launch event is today, but GitHub accidentally leaked the entire lineup, revealing four new models, including GPT-5 Mini and Nano.
OpenAI has confirmed that it will host an online launch event for its next foundational model GPT-5 on Thursday, and not long after that GitHub blew the lid off the AI brand's entire lineup.
The new family of GPT-5 models appeared on a since deleted post on GitHub, in an accidental leak Wednesday evening. The leak detailed that GPT-5 will be available in four versions, essentially confirming the tiered option rumor that has been circulating about the AI model for some time. There has also already been reports of mini and nano versions. The models are set to offer GPT-5 “major improvements in reasoning, code quality, and user experience,” according to the leaked blog.
The GPT-5 models featured on the GitHub blog include:
gpt-5: Designed for logic and multi-step tasks.
gpt-5-mini: A lightweight version for cost-sensitive applications.
gpt-5-nano: Optimized for speed and ideal for applications requiring low latency.
gpt-5-chat: Designed for advanced, natural, multimodal, and context-aware conversations for enterprise applications.
Users on Reddit discovered and explored details of the GitHub post and the models before the blog post was removed. However, the page was archived. The post also noted that GPT-5 will include “enhanced agentic capabilities” and will be able to process “complex coding tasks with minimal prompting.”
The GPT-5 launch announcement and leak have come immediately after OpenAI seemingly announced its open weight gpt-oss-120b and gpt-oss-20b models as a diversion away from the foundational model. Many rumors suggested that GPT-5 would be announced sometime in August, especially with OpenAI's CEO, Sam Altman sharing several teasers about the new AI model.
The GPT-5 launch event is streaming online at 1 p.m. ET/10 a.m. PT.