Freeplay
Freeplay is an LLM product experimentation, evaluation, and observability platform for cross-functional teams. Its HTTP API and SDKs make Freeplay the source of truth for prompt templates, record completions and sessions/traces from production, curate test datasets, run batch test runs and LLM-judge evaluations, and capture human and customer feedback.
APIs
Freeplay Prompt Templates API
Manage versioned prompt templates as the source of truth for an application, including creating templates and versions, retrieving formatted or raw templates by name or ID, depl...
Freeplay Recordings & Sessions API
Record LLM completions back to Freeplay along with the sessions and traces that group related calls for agent workflows, then list, search, and delete sessions and aggregate com...
Freeplay Test Cases & Datasets API
Curate evaluation datasets and their test cases, retrieving dataset metadata and test cases by name or ID and uploading new test cases for use in batch test runs and experiments.
Freeplay Test Runs & Evaluations API
Run batch evaluations against datasets, create and list test runs and retrieve their results, and record completion-level and trace-level customer and human feedback to close th...