Quick Start

you can start exploring amberflo immediately using the built in playground the playground lets you send prompts to real ai models using your own provider credentials every request is captured, metered, and automatically converted into cost so you can immediately understand how your ai usage translates into spend getting started takes just three steps set up credentials and models for your ai providers create a workload to track and attribute usage and cost generate a virtual key for the workload once these steps are complete, you can use the playground to send prompts to your configured models and begin seeing real usage and cost data inside amberflo step 1 set up providers & models first, connect the ai providers you want to use in the left hand navigation, go to model management add a provider select the providers tab click add provider in the upper right enter a name for the provider select the provider type (such as openai, azure openai, aws bedrock, or others) enter the required authentication details click save provider your provider credentials are stored securely and used when sending requests from the playground add models after adding a provider, configure the models you want to use in model management , click the models tab click add model select the provider you just created choose one of the models available from that provider optionally assign a model alias this makes models easier to reference click save amberflo automatically pulls in public pricing information for supported models this allows the platform to immediately calculate the cost of requests made through the playground step 2 create a workload next, create a workload a workload is the unit used to track and attribute ai usage and cost workloads can represent applications teams features customers environments to create a workload in the left navigation, go to access management → workloads click create workload enter a workload name review or edit the automatically generated workload id select the models this workload is allowed to access click create workload all usage generated through keys associated with this workload will roll up into the same usage, cost, and analytics views step 3 create a virtual key virtual keys allow controlled access to models through a specific workload to create one go to access management → keys click add key select the workload you created enter a name for the key click create key the key value will be displayed once copy and store it securely if it is lost, you will need to generate a new key use the playground to generate your first ai usage now you are ready to start using the playground the playground provides a built in chat interface where you can send prompts to your configured models using the virtual key associated with your workload each request you make through the playground will call the ai provider using your credentials capture usage data such as tokens and request metadata automatically calculate the cost using model pricing attribute the usage and cost to your workload within moments, you will begin seeing real usage and cost data appear in amberflo explore your ai spend after sending a few prompts through the playground, navigate to ai spend there you will begin to see cost attributed to your workload token usage across models spend by provider and model request and usage trends over time this is where the power of amberflo becomes clear the same usage data generated by your ai interactions can now be used to understand cost, optimize usage, and eventually power customer billing what to do next once you are comfortable using the playground, you can create additional workloads for teams, applications, or customers explore ai spend dashboards to analyze usage and cost configure custom rates for internal cost tracking or customer billing build pricing models based on real ai usage with amberflo, the same usage signals that power cost visibility can also power accurate, usage based billing for ai powered products