OpenAI, the company behind ChatGPT, has announced the release of new language processing models and updates to existing models to perform more comprehensive tasks than the previous preview model. The new models include:
What are the new embedding models?
OpenAI refers to the two new embedding models as text-embedding-3-small, the smaller of the two models, and the more powerful text-embedding-3-large model.
According to OpenAI, the text-embedding-3-small model “offers significant upgrades compared to its predecessor” released in December 2022. It provides “stronger performance,” is “more efficient,” and will be available at a reduced price (5 times less) compared to the previous generation model.
Enhanced GPT-3.5 Turbo Model
The free version of ChatGPT uses the GPT-3.5 Turbo model, and the company is making various enhancements, including greater accuracy in responding to suggested prompts and fixing an error that affected non-English text encoding. The company is also reducing the price of the model.
GPT-4 Turbo Preview Model
The company has announced that GPT-4 Turbo is the most powerful model derived from GPT-4, which was released last year. OpenAI reported that over 70% of requests from GPT-4 API customers have transitioned to GPT-4 Turbo from the initial release. The company has released an updated preview model of GPT-4 Turbo, which is supposed to complete tasks like code generation more systematically than the previous preview model. The new model also includes a fix for the error affecting the generation of non-English UTF-8 general text.
Frequently Asked Questions
1. What are the new embedding models?
The new embedding models announced by OpenAI are the text-embedding-3-small and text-embedding-3-large models. The former is smaller, and the latter is more powerful.
2. What are the enhancements of the text-embedding-3-small model?
The text-embedding-3-small model offers significant upgrades compared to its predecessor. It has stronger performance, is more efficient, and is available at a reduced price compared to the previous model.
3. What improvements are expected in the GPT-3.5 Turbo model?
The GPT-3.5 Turbo model is expected to have improvements such as greater accuracy in response, fixing an error related to non-English text encoding, and a reduction in price.
4. What is GPT-4 Turbo?
GPT-4 Turbo is the most powerful model derived from GPT-4, and OpenAI has released an updated preview model of GPT-4 Turbo. This new model has improved performance and fixes for errors that affect the generation of non-English general text.
OpenAI: A company that develops artificial intelligence and advanced language processing models.
ChatGPT: A model used for generating conversations or text based on user input.
GPT-3.5 Turbo: A language processing model used by ChatGPT to provide responses and information in suggested text prompts.
GPT-4 Turbo: An enhanced model of GPT-4 with improved performance and the ability to generate code and general text.