OpenAI Rolls Out GPT-4 Turbo, Cheaper GPT-3.5, New Embedding Models, and Content Moderation Upgrades
-
OpenAI released an updated GPT-4 Turbo preview model intended to reduce "laziness" where the AI doesn't thoroughly complete tasks.
-
New GPT-3.5 Turbo model announced with accuracy improvements, lower pricing (50% less for input, 25% less for output).
-
New embedding models, text-embedding-3-small and text-embedding-3-large, released to aid machine learning tasks.
-
Updated text moderation model, text-moderation-007, rolled out to help developers identify potentially harmful text.
-
Improvements to OpenAI's developer platform introduced like new tools for managing API keys and monitoring API usage.