OpenAI, a leading organization in the field of generative AI, has announced significant updates to its text-generating models while also reducing pricing. The company unveiled new versions of GPT-3.5-turbo and GPT-4, the latest addition to its text-generating AI lineup. Notably, GPT-4 introduces a feature called function calling, which allows developers to describe programming functions to the model and have it generate code to execute those functions.

Function calling opens up possibilities for various applications, such as creating chatbots that leverage external tools, converting natural language into database queries, and extracting structured data from text. The models have been fine-tuned to recognize when a function needs to be called and respond with JSON formatted according to the function’s signature. This enables developers to obtain structured data more reliably from the models.

In addition to function calling, OpenAI is enhancing GPT-3.5-turbo with an expanded context window. The context window refers to the amount of text the model considers before generating new content. By increasing the context length fourfold to 16,000 tokens, the model can retain more information from recent conversations, reducing the likelihood of veering off-topic or generating problematic responses.

While the expanded context window comes at a higher price of $0.003 per 1,000 input tokens and $0.004 per 1,000 output tokens, OpenAI has simultaneously reduced pricing for the original GPT-3.5-turbo by 25%. Developers can now utilize the model for $0.0015 per 1,000 input tokens and $0.002 per 1,000 output tokens, enabling approximately 700 pages of text per dollar.

OpenAI has also lowered the price of text-embedding-ada-002, a popular text embedding model. Text embeddings measure the relatedness of text strings and are widely used in search and recommendation systems. The cost of text-embedding-ada-002 has been reduced by 75% to $0.0001 per 1,000 tokens, thanks to increased system efficiency within OpenAI.

It’s worth noting that OpenAI has emphasized incremental updates to existing models rather than focusing on developing new models from scratch. CEO Sam Altman clarified that the successor to GPT-4 is yet to begin training, indicating the company’s dedication to refining and improving their current offerings. OpenAI continues to invest heavily in research and infrastructure, demonstrating its commitment to pushing the boundaries of generative AI technology.

OpenAI, a leading organization in the field of generative AI, has announced significant updates to its text-generating models while also reducing pricing. The company unveiled new versions of GPT-3.5-turbo and GPT-4, the latest addition to its text-generating AI lineup. Notably, GPT-4 introduces a feature called function calling, which allows developers to describe programming functions to the model and have it generate code to execute those functions.

Function calling opens up possibilities for various applications, such as creating chatbots that leverage external tools, converting natural language into database queries, and extracting structured data from text. The models have been fine-tuned to recognize when a function needs to be called and respond with JSON formatted according to the function’s signature. This enables developers to obtain structured data more reliably from the models.

In addition to function calling, OpenAI is enhancing GPT-3.5-turbo with an expanded context window. The context window refers to the amount of text the model considers before generating new content. By increasing the context length fourfold to 16,000 tokens, the model can retain more information from recent conversations, reducing the likelihood of veering off-topic or generating problematic responses.

While the expanded context window comes at a higher price of $0.003 per 1,000 input tokens and $0.004 per 1,000 output tokens, OpenAI has simultaneously reduced pricing for the original GPT-3.5-turbo by 25%. Developers can now utilize the model for $0.0015 per 1,000 input tokens and $0.002 per 1,000 output tokens, enabling approximately 700 pages of text per dollar.

OpenAI has also lowered the price of text-embedding-ada-002, a popular text embedding model. Text embeddings measure the relatedness of text strings and are widely used in search and recommendation systems. The cost of text-embedding-ada-002 has been reduced by 75% to $0.0001 per 1,000 tokens, thanks to increased system efficiency within OpenAI.

It’s worth noting that OpenAI has emphasized incremental updates to existing models rather than focusing on developing new models from scratch. CEO Sam Altman clarified that the successor to GPT-4 is yet to begin training, indicating the company’s dedication to refining and improving their current offerings. OpenAI continues to invest heavily in research and infrastructure, demonstrating its commitment to pushing the boundaries of generative AI technology.