OpenAI announces free fine-tuning for GPT-4o mini model

OpenAI is offering free fine-tuning on its new GPT-4o mini model, allowing users to train the model on additional data at no charge to enable higher performance for specific use cases.

GPT-4o mini fine-tuning is available to developers in OpenAI’s Tier 4 and 5 usage tiers, which are the highest-priced tiers among OpenAI’s plans. OpenAI plans to gradually expand access to free fine-tuning to all tiers. Free fine-tuning will be offered now through September 23.

Developers can start fine-tuning GPT-4o mini for free by visiting their fine-tuning dashboard, clicking “create,” and selecting “GPT-4o mini” from the base model drop-down menu. Each organization gets 2M training tokens per 24-hour period to train the model. Any overage will be charge $3.00 per 1M tokens. More details on free fine-tuning are offered in OpenAI’s fine-tuning docs.

Compared to fine-tuning with OpenAI’s GPT 3.5 Turbo, GPT-4o mini is positioned to be more cost-efficient and more capable, with longer context and lower latency. GPT-4o mini was launched July 18 with the intention of expanding the range of applications built with AI by making intelligence more affordable. It enables a range of tasks such as applications that chain or parallelize model calls, pass a large volume of context to the model, or interact with customers through fast real-time text responses, such as customer support chatbots.