Your prompts are not used as training data at ZC Technologies. We operate on a privacy-first model, ensuring that all interactions with our Qwen 2.5 models are handled with the utmost confidentiality. This is particularly important for developers in regulated industries such as legal, healthcare, and finance, where data privacy and security are paramount.
At ZC Technologies, we do not log prompts or responses for model training. Instead, we use owned hardware, specifically dedicated NVIDIA GB10 silicon, to run our Qwen 2.5 models. This setup ensures that your data remains within your organization's control and is not shared with third parties or used to train our models.
Our ZC Inference Exchange platform supports both Qwen 2.5 32B and 72B models, providing a robust solution for a variety of use cases. The platform is accessible via an OpenAI-compatible API endpoint at /v1/chat, with authentication handled through Bearer tokens. This ensures that only authorized users can access the models, further enhancing security.
Our pricing is designed to be competitive, undercutting major competitors by 60-80% per 1M tokens. Here are the current subscription plans:
By using our platform, you can ensure that your sensitive data remains secure and compliant with industry regulations. Our privacy-first approach means that your prompts and responses are not used to train our models, which is a critical feature for regulated industries.
To get started with ZC Technologies, sign up for a prepaid LLM credit line at https://zcx.zctechnologies.org#plans. Choose the plan that best fits your needs, and begin leveraging our secure and efficient AI models today.
# Example of making an API call
POST https://zcx.zctechnologies.org/v1/chat
Authorization: Bearer YOUR_BEARER_TOKEN
Content-Type: application/json
{
"model": "qwen2.5:72b",
"messages": [
{"role": "user", "content": "What is the capital of France?"}
]
}For more information on our privacy model and how we can support your development needs, visit https://zcx.zctechnologies.org#plans.