Microsoft GitHub today announced the preview availability of two new OpenAI models, o1-preview and o1-mini, in both GitHub Copilot and Models. OpenAI launched the new o1 series of AI models designed to spend more time thinking before they respond. These models can reason through complex tasks and solve harder problems than previous OpenAI models in science, coding, and math.
When using o1-preview with GitHub Copilot, the GitHub team noticed that the model’s reasoning capability allows for a deeper understanding of code constraints and edge cases, producing a more efficient and higher quality result. With this new preview availability, developers can select either o1-preview or o1-mini to power their Copilot Chat experience in VS Code instead of the current default model, GPT-4o. GitHub will also allow them to change models during a conversation based on the requirement. For example, developers can use GPT-4o to get a quick explanation of APIs and use o1-preview for designing complex algorithms.
Developers can also test both new o1 models in the playground in GitHub Models to explore their capabilities and performance. Later, they can integrate the models into their apps for further development.
Developers can now sign up here to get access to the new OpenAI o1 via GitHub Copilot Chat with Visual Studio Code and/or via GitHub Models playground.
On a related note, OpenAI recently increased the rate limits of the o1-preview and o1-mini APIs for developers. The o1-preview API now supports 500 requests per minute, while the o1-mini API now supports 1000 requests per minute. The OpenAI team is working further to increase the rate limits and expand the API access to more developer tiers.
On the consumer side, OpenAI o1-preview and o1-mini models are now available to all ChatGPT Enterprise and ChatGPT Edu customers. OpenAI recently increased rate limits for o1-mini by 7x, from 50 messages per week to 50 messages per day. And the rate limit for o1-preview has been increased from 30 messages per week to 50 messages per week.
Source: GitHub