Frequently Asked Questions

What is the roadmap for the product?
How is pricing working?
How do you estimate the cost savings?
What does tensormesh consider a cache hit?
How long before I can access the beta?
What if my model is not offered, can I still use it?
What if I want to use another GPU provider?
Can I use Ternsormesh on my own hardware?
Can I share the cache between multiple servers in a cluster?
Are incoming queries routed to the right node based on cache affinity?
Can I use a different inference server?
Can I perform the same action I can do on the Web UI through an API or a CLI?
For security reasons, I need to be hosting my own inference stack, is Tensormesh available for an on-prem deployment?