OpenAI launches GPT-4o mini fine-tuning function with 2 million free training tokens every day

OpenAI recently announced the launch of fine-tuning functions for its new small-scale model GPT-4o mini. This initiative is designed to improve the performance of the model in specific application scenarios and provide developers and enterprise users with more customized AI solutions. This update will lower the threshold for AI applications and bring more possibilities to the popularization and application of AI technology. Compared with the previous generation model, GPT-4o mini has significantly improved its cost performance, context length and capabilities. The opening of its fine-tuning function also marks OpenAI's determination to continue to innovate in the field of AI.

Recently, OpenAI, a leading artificial intelligence company, announced the launch of fine-tuning functions for its new small-scale model GPT-4o mini, aiming to improve the performance of the model in specific application scenarios. This move will bring more customized AI solutions to developers and enterprise users.

According to official news from OpenAI, from now until September 23, users can fine-tune GPT-4o mini for free every day, with up to 2 million training tokens. Any excess will be charged at USD 3.00 per million Tokens. After September 24, all fine-tuning training will be billed at a uniform rate of US$3.00 per million Tokens.

OpenAI highlighted several advantages of GPT-4o mini over GPT-3.5 Turbo:

1. More cost-effective: GPT-4o mini’s input token fee is 90% lower than GPT-3.5 Turbo’s, and output token fee is 80% lower. Even after the free period ends, its training cost is still 50% lower than GPT-3.5 Turbo.
2. Longer context: The training context length of GPT-4o mini is 65k Tokens, 4 times that of GPT-3.5 Turbo; the inference context length is 128k Tokens, 8 times that of GPT-3.5 Turbo.
3. More powerful capabilities: OpenAI stated that GPT-4o mini is not only more intelligent than GPT-3.5 Turbo, but also supports visual functions (currently fine-tuning is limited to text).

This fine-tuning feature is available to enterprise customers and developers in Tier 4 and Tier 5 usage levels. Users can start free fine-tuning by accessing the fine-tuning dashboard and selecting gpt-4o-mini-2024-07-18 as the base model.

OpenAI also provides users with detailed fine-tuning documentation and a "How to fine-tune the chat model" guide to ensure that users can make full use of this new feature. If you have any questions, users can seek help in the OpenAI Developer Forum.

Industry experts believe that the launch of GPT-4o mini and the opening of fine-tuning functions will further lower the threshold for AI applications and promote the birth of more innovative applications. This is not only conducive to the prosperity of the developer community, but will also provide more possibilities for the implementation of AI in various industries.

This move by OpenAI once again demonstrates its leading position in the field of AI and also reflects the company's determination to make AI technology more popular and easier to use. With the promotion of fine-tuning functions, we have reason to expect to see more innovative applications based on GPT-4o mini emerging in the near future.

All in all, the launch of GPT-4o mini and the opening of its fine-tuning function herald the further popularization and development of AI technology applications, providing developers and enterprises with more convenient, cost-effective AI solutions. It is worth looking forward to its future application in various industries. Wide range of applications in various industries.