Home InternationalIntroducing improvements to the fine-tuning API an...
International⭐ Featured

Introducing improvements to the fine-tuning API and expanding our custom models program

We’re adding new features to help developers have more control over fine-tuning and announcing new ways to build custom models with OpenAI.

6 April 2026 at 12:58 pm
1 views
Introducing improvements to the fine-tuning API and expanding our custom models program

OpenAI has announced significant updates to its fine-tuning API and an expansion of its custom models program, offering developers greater control and flexibility in tailoring AI solutions to their specific needs. These improvements aim to enhance the capabilities of developers working with OpenAI's models, enabling them to fine-tune and customize these models more effectively.

The new features added to the fine-tuning API provide developers with enhanced control over the fine-tuning process. This includes more granular options for adjusting hyperparameters, such as learning rates and batch sizes, allowing for more precise optimization of model performance. Additionally, the API now supports a wider range of customization options, including the ability to specify custom tokenizers and preprocessing pipelines. These updates enable developers to fine-tune models that better align with their specific datasets and use cases, leading to improved accuracy and efficiency.

Another key development is the expansion of OpenAI's custom models program. This initiative allows developers to build and train models tailored to their unique requirements, leveraging OpenAI's infrastructure and expertise. The program now offers new ways to collaborate with OpenAI's team of AI experts, providing guidance and support throughout the model development process. This partnership enables developers to harness the full potential of custom models, ensuring they are optimized for their specific applications.

OpenAI has also introduced new tools and resources to facilitate the development of custom models. These include detailed documentation, interactive tutorials, and access to a growing library of pre-trained models that can serve as a starting point for customization. By providing these resources, OpenAI empowers developers to experiment with different model architectures and configurations, ultimately leading to the creation of highly specialized AI solutions.

These updates to the fine-tuning API and the custom models program reflect OpenAI's commitment to empowering developers with the tools they need to build innovative AI applications. By offering greater control and flexibility, OpenAI is enabling a new wave of AI-driven solutions that are more effective and efficient than ever before.

In the coming months, OpenAI plans to continue enhancing its offerings, adding more features and capabilities to support developers in their AI endeavors. With these improvements, developers can expect even greater flexibility and control over their AI projects, allowing them to achieve their goals more efficiently and effectively.

In conclusion, OpenAI's recent updates to its fine-tuning API and custom models program represent a significant step forward in the field of AI development. By providing developers with more control and flexibility, OpenAI is enabling a new generation of AI solutions that are tailored to the specific needs of businesses and individuals alike. As OpenAI continues to innovate and expand its offerings, the possibilities for what can be achieved with AI are only growing.

Source: OpenAI News
📰 Related News
Ollama 0.2.6 Released with Native Gemma 4 Support and Enhanced Performance
Ollama 0.2.6 Released with Native Gemma 4 Support and Enhanced Performance
Ollama 0.2.6 is now live, featuring native support for Google's Gemma 4 models and improved local inference performance for Windows, macOS, and Linux.
14 Apr
Weekly news roundup: Shortages spread to MLCCs; SK Hynix reportedly in talks with Microsoft and Google
Weekly news roundup: Shortages spread to MLCCs; SK Hynix reportedly in talks with Microsoft and Google
Below are the most-read DIGITIMES Asia stories from the week of April 6-April 13, 2026:
14 Apr
cutile-stencil 0.2.0
cutile-stencil 0.2.0
An xDSL-based stencil compiler that generates optimized GPU kernels via NVIDIA cuTile
14 Apr
merlin-llm added to PyPI
merlin-llm added to PyPI
Merlin — a fast local LLM for agentic coding on Apple Silicon
14 Apr
Fluent Cut - Craft and compose videos programmatically in PHP with an elegant fluent API
Fluent Cut - Craft and compose videos programmatically in PHP with an elegant fluent API
Craft and compose videos programmatically in PHP with an elegant fluent API - b7s/fluentcut
14 Apr
Crypto Investor at Center of Trump Corruption Allegations Now Sees Himself as ‘Victim’
Crypto Investor at Center of Trump Corruption Allegations Now Sees Himself as ‘Victim’
Justin Sun has accused Trump-affiliated World Liberty Financial of misconduct and a general lack of transparency.
14 Apr
nvidia-nat-weave 1.7.0a20260413
nvidia-nat-weave 1.7.0a20260413
Subpackage for Weave integration in NeMo Agent Toolkit
14 Apr
nvidia-nat-s3 1.7.0a20260413
nvidia-nat-s3 1.7.0a20260413
Subpackage for S3-compatible integration in NeMo Agent Toolkit
14 Apr
Social Security Trust Fund to Run Dry in 2032: Just 6 Years From Now
Social Security Trust Fund to Run Dry in 2032: Just 6 Years From Now
Six years. That is how much time separates retirees from a Social Security system that, by its own projections, runs out of money. If you are 56 years old...
14 Apr
cane-gpu-perf added to PyPI
cane-gpu-perf added to PyPI
GPU inference benchmarking with opinionated diagnostics
13 Apr