Asynchronous Model Execution With Concurrent Request Handling

CLI tool for interacting with LLMs.

Unique: Provides parallel sync and async class hierarchies (Model/AsyncModel, KeyModel/AsyncKeyModel) allowing developers to choose the execution model that fits their application. The async API is identical to the sync API, just with async/await syntax, minimizing the learning curve.

vs others: More integrated than manually wrapping sync calls with asyncio.to_thread because async is built into the model abstraction; more efficient than thread-based concurrency because it avoids thread overhead; simpler than building custom async wrappers because the abstraction handles provider-specific async implementations.