OpenAI's Latest Advanced LLM Model, Orion, to Be Released by December: Potentially 100 Times More Powerful Than GPT-4
OpenAI is gearing up to release its next-generation large language model (LLM), codenamed Orion, by December. This model is expected to be significantly more powerful than GPT-4, potentially up to 100 times stronger, according to Verge.
Unlike previous models, OpenAI plans to grant initial access to Orion to select partner companies, allowing them to develop their own products and features. This strategy is aimed at fostering collaboration with organizations closely aligned with OpenAI's goals. Microsoft, OpenAI's major investor, is reportedly preparing to host Orion on its Azure platform as early as November.
While Orion is internally considered the successor to GPT-4, it's unclear if OpenAI will officially call it GPT-5. The company has been using a model codenamed Strawberry to generate synthetic data for training Orion, and the training process was reportedly completed in September.
The release of Orion marks a significant step for OpenAI, which recently secured $6.6 billion in its latest funding round, achieving a valuation of $157 billion. The company is restructuring itself as a for-profit entity and may be positioning itself for a future IPO.
OpenAI's long-term goal is to integrate its LLMs to create an even more capable model that could eventually be recognized as artificial general intelligence (AGI).
CEO Sam Altman on September just posted "excited for the winter constellations to rise soon." If ask ChatGPT o1-preview what Altman's post is hiding, it will tell you that he's hinting at the word Orion, which is the winter constellation that's most visible in the night sky from November to February.
However, OpenAI faces challenges, including the recent departure of several top executives, including the head of its AGI Readiness team. This team was responsible for advising OpenAI on its capacity to handle increasingly powerful AI and its readiness for the world. The company also saw CTO Mira Murati, research VP Barret Zoph, and research chief Bob McGrew left the AI company on the same day in September.