Skip to content

Releases: zkywsg/ModelConfig

v0.1.0

15 Mar 02:54
@zkywsg zkywsg
v0.1.0
126578f

Choose a tag to compare

View all tags
v0.1.0 Latest
Latest

ModelConfig v0.1.0 is the first public release.

ModelConfig is a resolver-only configuration layer for multi-provider AI apps. It lets application code depend on logical models like smart or assistant, then resolves them into concrete provider targets at runtime.

Highlights:

  • logical model routing
  • provider/model decoupling
  • alias resolution
  • capability-aware selection
  • environment-specific policies
  • OpenAI SDK example
  • LiteLLM example
  • JSON Schema-backed config

This release is focused on a narrow boundary:
ModelConfig decides which target to use. Your SDK, gateway, or client executes the request.

If you are building multi-provider AI systems and hit routing edge cases, open an issue with the scenario.

Assets 2
Loading