OpenAI Launches Foundry: A Platform for Running Machine Learning Models

OpenAI Launches Foundry: A Platform for Running Machine Learning Models
OpenAI Launches Foundry: A Platform for Running Machine Learning Models


OpenAI has announced the launch of Foundry, a new developer platform for running its newer machine learning models like GPT-3.5 on dedicated capacity. Foundry is designed for cutting-edge customers running larger workloads. According to screenshots of documentation published on Twitter by users with early access, Foundry allows inference at scale with full control over the model configuration and performance profile. The platform will offer service-level commitments for instance uptime and on-calendar engineering support.

Features of Foundry

Foundry offers the following features:

Static Allocation of Compute Capacity: Foundry delivers a “static allocation” of compute capacity on Azure, OpenAI’s preferred public cloud platform dedicated to a single customer.

Version Control: Foundry provides some level of version control, letting customers decide whether or not to upgrade to newer model releases, as well as “more robust” fine-tuning for OpenAI’s latest models.

Monitoring Capabilities: Users will be able to monitor specific instances with the same tools and dashboards that OpenAI uses to build and optimize models.

Rentals based on dedicated compute units: Rentals will be based on dedicated compute units with three-month or one-year commitments. Running an individual model instance will require a specific number of compute units.

Pricing: Instances won’t be cheap. Running a lightweight version of GPT-3.5 will cost $78,000 for a three-month commitment or $264,000 over a one-year commitment.

Implications of Foundry

Foundry is expected to be a game-changer for customers running large machine learning workloads. It will allow customers to run OpenAI's latest models with greater control, customization, and performance. Foundry also represents a significant move towards monetization for OpenAI, which has been under pressure to turn a profit after a multibillion-dollar investment from Microsoft. According to reports, the company expects to make $200 million in 2023. Compute costs have been a major hurdle for the company, and Foundry will help OpenAI generate revenue by charging customers for its dedicated compute units.

Possible GPT-4 Launch?

Eagle-eyed Twitter and Reddit users spotted that one of the text-generating models listed in the instance pricing chart has a 32k max context window. GPT-3.5, OpenAI’s latest text-generating model, has a 4k max context window, suggesting that this mysterious new model could be the long-awaited GPT-4 — or a stepping stone toward it.

Conclusion

OpenAI's Foundry platform will enable customers to run its latest machine learning models with greater control and customization. Foundry will help OpenAI generate revenue by charging customers for its dedicated compute units. With the possibility of a GPT-4 launch, OpenAI's tech is expected to play a major role in the future of artificial intelligence.

Comments

Post a Comment

Popular posts from this blog

AI 'godfather' Geoffrey Hinton warns of dangers as he quits Google

Image
Geoffrey Hinton , a renowned pioneer in artificial intelligence (AI) and a former leader of Google’s AI research division, has resigned from his position at the tech giant. Hinton has cited growing concerns about the ethical implications of the technology he helped create as his reason for leaving. He aims to speak more openly about the potential risks and harms of AI , especially as the company and its rivals have been racing to develop and deploy ever more powerful and sophisticated models. Hinton’s departure is the latest and most prominent sign of a growing rift between some of the world’s leading AI researchers and the tech companies that employ them. Hinton is widely regarded as the “ Godfather of AI ” for his groundbreaking work on deep learning and neural networks. He expressed his concerns about the impact of AI on society, and he has warned about the existential threat of superintelligent AI that could surpass human capabilities and goals. In an interview with the New York Ti...
Read more

OpenAI's Bug Bounty Program to Find Bugs in ChatGPT

Image
OpenAI, the company behind the artificial intelligence chatbot ChatGPT, has announced a new program to reward users who identify and report software bugs in its system. The company is inviting security researchers, ethical hackers, and technology enthusiasts to help identify and address vulnerabilities in ChatGPT, OpenAI’s plugins, the OpenAI API, and other related services. Rewards will range from $200 for low-severity findings to up to $20,000 for exceptional discoveries , based on the severity and impact of the reported issues. OpenAI has partnered with the bug bounty platform Bugcrowd to streamline the submission and reward process. The company has also released guidelines and rules of engagement, stating what will not be rewarded, such as getting the AI model to say bad things or writing malicious code. OpenAI has urged users who find bugs to report the vulnerabilities promptly and unconditionally and not engage in extortion, threats, or other tactics to elicit a response under d...
Read more

Microsoft Expands Partnership with OpenAI

Image
Microsoft and OpenAI have announced an expansion of their partnership to bring advanced AI capabilities to more developers and businesses. The partnership will focus on providing easy access to OpenAI's powerful GPT-3 language model through Microsoft's Azure platform , allowing developers to easily integrate the model into their applications and services. One of the key components of the partnership is the development of an Azure-hosted GPT-3 model , which will make it easy for developers to access the model's capabilities without the need for expensive hardware and infrastructure. This will allow developers to easily add natural language processing capabilities to their applications, such as text generation, language translation, and question answering. Another aspect of the partnership is the integration of GPT-3 into Azure Cognitive Services , which will allow developers to use the model's capabilities through a simple API. This will make it easy for developers ...
Read more