OpenAI unveils GPT-4 Turbo, its most powerful AI to date

OpenAI has introduced GPT-4 Turbo, an upgraded version of its text-generating AI model, GPT-4.

This improved model is claimed to be more powerful and efficient than its predecessor. GPT-4 Turbo offers two versions: one for text analysis and one for both text and image processing. It is available in preview via an API, with plans for a general release in the coming weeks.

Pricing for GPT-4 Turbo is set at $0.01 per 1,000 input tokens and $0.03 per 1,000 output tokens. The cost for image processing depends on the image size. OpenAI claims that GPT-4 Turbo is three times cheaper for input tokens and twice as cheap for output tokens compared to GPT-4.

Key improvements in GPT-4 Turbo include a more recent knowledge base, with a cut-off date in April 2023, enabling it to provide more accurate responses to recent events. It also boasts a substantial context window, allowing it to remember and reference longer text segments.

Additionally, it supports a new "JSON mode" and provides improved performance in tasks that require precise instructions.

OpenAI is also launching an experimental fine-tuning program for GPT-4, though it will involve more oversight and guidance due to technical complexities. The company has increased the tokens-per-minute rate limit for GPT-4 customers without changing the pricing structure.