Corvex Launches Security-First AI Inference Platform for Open-Weight Models

Corvex (MOVE) launched Corvex Token Factory, a platform that lets developers and enterprises access open-weight AI models through an API.

Rhea-AI Summary

Corvex (MOVE) launched Corvex Token Factory, a platform that lets developers and enterprises access open-weight AI models through an API. Customers do not need to deploy or operate GPU clusters. Initial models include GLM 5.3 from Z.ai and DeepSeek V4 Flash 0731.

Zero data retention is the default: prompts and responses are processed in memory, not stored or used for training. Corvex retains limited operational metadata that excludes prompt and response content. Requests run on Corvex-managed hardware without routing to third-party inference providers or model developers. Customers pay for input and output tokens used, without idle GPU charges. Feature compatibility varies by client and model.

News Explained

Corvex says it signs Business Associate Agreements with customers subject to HIPAA, specifying a contractual arrangement for that customer group alongside the launched service.

Key Terms

inference, zero data retention, api, soc 2 type ii, +1 more

5 terms

inference technical

"Inference is the process that turns an end-user request into a model-generated response."

Inference is the process of drawing a conclusion from available evidence or data, like a detective piecing together clues to form a likely story. For investors it matters because these judgments turn raw reports, test results, or market signals into expectations about future performance, risk, or regulatory outcomes—so how someone infers from the same facts can change investment decisions and valuation.

zero data retention technical

"At launch, Corvex Token Factory is built around four principles: Assurance – Zero data retention is the default."

A data handling policy or technical configuration under which collected personal or transaction data is not stored persistently after it has been processed; data exists only transiently (for the duration of a session or in volatile memory) and is deleted or overwritten immediately once its immediate purpose is complete. Zero data retention does not necessarily mean nothing is ever transmitted or temporarily buffered, and it can be constrained by other systems such as backups, logs, audit records, or legal and regulatory obligations that require keeping certain information.

api technical

"Corvex Token Factory gives customers access to open-weight AI models through an API"

An API, or Application Programming Interface, is a set of rules that allows different software programs to communicate and work together smoothly, much like a waiter translating your order into the kitchen and then bringing your meal back. For investors, APIs are important because they enable real-time access to financial data, trading systems, and other digital services, making it easier to make informed decisions quickly and efficiently.

soc 2 type ii regulatory

"The Company is SOC 2 Type II certified."

A SOC 2 Type II report is an independent audit that verifies a company’s operational controls for protecting customer data work effectively over a sustained period, not just at a single point in time. Think of it like a multi-month health inspection for a business’s data practices: it reassures customers and investors that systems are managed reliably, lowering the risk of breaches, service interruptions, and regulatory or liability costs.

hipaa regulatory

"customers that are subject to HIPAA"

A U.S. law that sets rules for keeping individuals’ health information private and secure, and for how that information can be shared. Think of it as a mandatory lock-and-key system for medical records that hospitals, insurers, and tech vendors must use. Investors care because failing to follow these rules can lead to big fines, costly remediation, loss of business access to patient data, and reputational damage that can hurt a company’s finances and growth prospects.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

Corvex Token Factory combines zero data retention by default with API-based access

ARLINGTON, Va., Sept. 30, 2026 /PRNewswire/ -- Corvex, Inc. (Nasdaq: MOVE), an engineering-led AI computing company, today announced the launch of Corvex Token Factory, its serverless inference platform built to help developers and enterprises use open-weight AI models with greater control over their data.

Corvex Token Factory gives customers access to open-weight AI models through an API without requiring them to deploy or operate GPU clusters. Initial model availability includes GLM 5.3 from Z.ai and DeepSeek V4 Flash 0731. The launch extends Corvex's reach to developers and enterprise technology teams that want to use AI as a service.

"As companies put AI to work on their source code, customer records and internal documents, they need to know where that information goes and what happens to it," said Seth Demsey, co-CEO and co-founder of Corvex. "With Corvex Token Factory, by default, we do not store end-user prompts or responses nor do we train on any user data. We have built these controls into a service that developers can easily use with their existing coding harnesses and AI API calls so that they can safely enjoy the power of leading open-weights models at a fraction of the cost of closed-source alternatives."

Inference is the process that turns an end-user request into a model-generated response. Corvex runs every request on hardware it manages, and it operates the models itself. Requests are never routed to third-party inference providers or forwarded to the model developers.

At launch, Corvex Token Factory is built around four principles:

• Assurance – Zero data retention is the default. Prompts and responses are processed in memory, never logged or stored, and never used to train models. Corvex retains limited operational metadata for security, service operations and billing, but that metadata does not include prompt or response content.
• Reliability – Corvex Token Factory is engineered to hold up under sustained workloads so it runs as a dependable execution layer in production workflows.
• Performance – Corvex tunes its inference stack, balancing first response and throughput under load. This supports interactive applications and agent workflows.
• Simplicity – OpenAI- and Anthropic-compatible APIs let developers connect existing clients by changing the base URL, API key and model name. Corvex manages the underlying infrastructure; feature compatibility varies by client and model.

Corvex is SOC 2 Type II certified. The Company also signs a Business Associate Agreement with customers that are subject to HIPAA, with security controls documented in the Corvex Trust Center.

Customers pay for the input and output tokens they use, without idle GPU charges. The Corvex Token Factory also provides a starting point for deeper customer relationships. Corvex's roadmap includes dedicated enterprise deployments and secure private inference environments, giving customers a path to adopt additional services as their usage and requirements grow.

The launch follows Corvex Token Factory's closed alpha period. Customers can register at no cost at https://tokenfactory.corvex.cloud/app and get an API key. Product documentation is available at https://docs.corvex.cloud/.

About Corvex, Inc.

Corvex, Inc. (Nasdaq: MOVE) is an AI cloud computing company specializing in GPU-accelerated infrastructure for AI workloads, and a publicly traded pure-play AI compute platform. The company provides secure, scalable and cost-efficient compute resources through GPU-accelerated clusters, high-throughput storage and a layered architecture engineered for reliability, performance and efficiency at scale. Corvex's product suite includes AI Factories and GPU Clusters, the Assured AI confidential-computing platform, as well as the Corvex Token Factory, a serverless inference platform. For more information, visit corvex.ai.

Cautionary Statements Regarding Forward-Looking Statements

This press release contains forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995 that are based upon current expectations or beliefs, as well as assumptions about future events. Forward-looking statements include all statements that are not historical facts and can generally be identified by terms such as "anticipate," "believe," "could," "estimate," "expect," "intend," "may," "plan," "potentially," "seek," "should," "target," "will" or "would" or similar expressions and the negatives of those terms. These statements include, but are not limited to, statements relating to product capabilities, customer deployment, business strategy, growth plans and objectives for future operations, anticipated power and customer-readiness dates, expected megawatt capacity and expansion options, and statements regarding anticipated customer offtake discussions, contract duration, pricing and attach opportunities. Actual results could differ materially as a result of, among other factors: the Company's inability to enter into customer offtake agreements at all, or on the duration, pricing or take-or-pay terms it anticipates; changes in prevailing market pricing for GPU-as-a-Service capacity; delays in GPU and long-lead equipment delivery; delays in utility energization or interconnection; local zoning, permitting and moratorium risk affecting expansion phases at either site; the Company's dependence on third-party site partners and the execution of definitive agreements with them; the Company's ability to obtain or maintain partner certifications and co-selling support; and the Company's limited operating history. Actual results could differ materially from those expressed in or implied by the forward-looking statements due to a number of risks and uncertainties, including the risks and uncertainties described in the Company's SEC reports, and under the heading "Risk Factors" in its most recent Annual Report on Form 10-K and Quarterly Reports on Form 10-Q, which are available at www.sec.gov and in other filings the Company makes and will make with the SEC. The forward-looking statements contained herein speak only as of the date of this press release. Except as required by law, the Company does not undertake any obligation to update or revise its forward-looking statements to reflect events or circumstances after the date of this press release.

Media Contact

Chris Donahoe, Stillpoint

corvex.media@stillpointglobaladvisors.com

 

View original content to download multimedia:https://www.prnewswire.com/news-releases/corvex-launches-security-first-ai-inference-platform-for-open-weight-models-302894497.html

SOURCE Corvex