Key Takeaways
- According to reports, Google plans to release Gemini 2.0 in December 2024, but performance expectations might not be met.
- It seems that businesses creating sophisticated AI models frequently experience this underperformance change.
- Gemini 2.0, which aims to give developers and users simultaneous access, is anticipated to replace Gemini 1.5.
According to reports, Google plans to release Gemini 2.0 in December 2024, although it might fall short of performance standards, which is a common occurrence with sophisticated AI models.
Google plans to release Gemini 2.0 in December, but the model has not achieved the expected performance gains, according to sources The Verge. The survey also revealed that businesses creating sophisticated models are familiar with this trend.
The multimodal big language models Gemini Ultra, Pro, Flash, and Nano are part of Google’s Gemini 1.0, which was released last December. Prior to rebranding, Bard offered the 1.0 Pro version, and the Nano was released concurrently with the Pixel 8 Pro. The 1.0 Ultra version was made available by Gemini Advanced in February 2024. Gemini 1.5, which featured an improved context window and nwas made available to subscribers in May, was also released in that month.
The “wide release” may indicate that Google intends to provide simultaneous access for programmers and customers, which could boost adoption and feedback. This new model will replace Gemini 1.5. In the meantime, Project Astra’s debut is still ongoing. It aims to improve Gemini Live with sophisticated imaging features by using the camera and microphone on your phone to boost context awareness. How Google will handle the timing of both initiatives is called into question by this.
According to a report from The Verge yesterday, OpenAI plans to introduce Orion—which is essentially ChatGPT-5—by the end of December. CEO Sam Altman emphasized, however, that this assertion is untrue. However, Orion’s initial rollout is expected to be more limited, involving selective partners.
