Corvex Launches Token Factory for Open Weight AI Inference

Sep 30, 2026
Corvex Token Factory provides API access to open weight AI models with zero data retention by default and usage based token pricing.

Corvex, Inc. launches Corvex Token Factory, a serverless inference service for open weight AI models, as it announced in a press release. The service initially supports GLM 5.3 from Z.ai and DeepSeek V4 Flash 0731 through an API.

Corvex says the service processes prompts and responses in memory without logging, storing, or using them for model training. Requests run on hardware managed by Corvex and are not sent to model developers or third party inference providers. The company retains operational metadata for security, billing, and service operations, but says this excludes prompt and response content.

The service offers OpenAI and Anthropic compatible APIs, allowing developers to connect existing clients by changing the base URL, API key, and model name. Customers pay for input and output tokens without charges for idle GPU capacity. Registration is available at no cost following a closed alpha period.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.