OpenAI new models update January 2024
![Paul Omenaca](/_next/image?url=%2Favatars%2Fpablo.jpeg&w=64&q=75)
Paul Omenaca
@houmland
![OpenAI new models update January 2024](/_next/image?url=%2Fblog-assets%2Fopenai-models-update-jan-2024%2Fopenai-new-models.png&w=1920&q=75)
Discover OpenAI's Latest Models on Stack AI
Stack AI introduces the latest advancements from OpenAI, bringing new models to our platform. In the rapidly evolving landscape of AI, staying up-to-date is important to leverage the full potential of Generative AI.
Introducing the Enhanced GPT-4 Turbo-Preview Model
OpenAI's new gpt-4-0125-preview is now available in Stack AI. As the most powerful model to date from OpenAI, it offers 128,000 tokens of context (i.e., unchanged from previous version), improving code generation and reducing the tendency of task 'laziness' - an issue where the model fails to complete tasks. This new model also includes a fix for a specific bug affecting non-english UTF-8 generations.
You can easily integrate this new model into your applications by dragging and dropping an OpenAI node and selecting gpt-4-turbo-Preview.
From now on, this model will always point to the latest version available, so you don't have to change it manually.
New Embedding Models: Raising the Bar for Data Retrieval
OpenAI has also released new embedding models (finally!), significantly enhancing the large language model's accuracy and context understanding.
The new text-embedding-3-small model significantly outperforms its predecessor, text-embedding-ada-002, with higher scores in multi-language retrieval and English task benchmarks.
Additionally, a larger model, text-embedding-3-large, is introduced, offering even higher performance with increased dimensions in embeddings.
These models are readily available in the 'Documents + Search' node settings.
Empowering Your Organization with different AI Models
At Stack AI, we maintain direct contact with leading generative AI providers, ensuring early access to the newest models as they emerge. This commitment positions us at the forefront of offering a model-agnostic, up-to-date generative AI platform.
We recognize the unique value each model brings and believe in harnessing their combined strengths to optimize outcomes. Our platform is designed with versatility in mind, enabling you to easily integrate and experiment with different models to find the best fit for your specific applications.
If you're interested in exploring how these diverse AI models can enhance your organization's capabilities, don't hesitate to contact us. Click here to schedule a meeting with us.