Introduction
Maximizing Speed, Minimizing Token Costs
Last updated
Maximizing Speed, Minimizing Token Costs
Last updated
Welcome to the cortecs docs! cortecs makes it easy to run dedicated language models at maximum performance.
Dedicated inference offers exclusive access to a specific model, ensuring that you are the sole user of the underlying compute resources. This makes it particularly suitable for applications that:
Need guaranteed latency
Have a heavy workload
Require many requests (no request limits)
Require high data security
In some use cases, like batched or scheduled jobs, it is useful to start the compute resources and shut them down automatically after the job is finished. Dynamic provisioning allows you to do exactly that - start an instance of the desired model, execute the job requiring LLM resources and shut it down when the resources are no longer required. That way, you are paying for the exact amount of resources you used, and not a minute more!
For examples see cortecs-py.
cortecs offers a variety of popular models. Visit our models page to explore the available options. Generally, more complex tasks require larger models, while smaller models provide faster performance. For most use cases, we recommend models supporting Instant provisioning.
Don't see a model you want to use? Join our Discord to add or upvote the model you'd love to use.
Register at cortecs.ai
Follow the quick start
Accomplish complex tasks using cortecs-py