Expand description
Model configuration and provider compatibility flags.
Structs§
- Cost
Config - Cost per million tokens (input/output).
- Model
Config - Full model configuration. Knows everything needed to make API calls.
- Open
AiCompat - Compatibility flags for OpenAI-compatible providers. Different providers have different quirks even though they share the same base API.
- Static
Credential Provider - Reference implementation of
CredentialProviderthat always returns a fixed key.
Enums§
- ApiProtocol
- Which API protocol a model uses.
- MaxTokens
Field - How a provider handles the
max_tokensfield. - Thinking
Format - How a provider formats thinking/reasoning output.
Traits§
- Credential
Provider - Pluggable source of the API key for a
ModelConfig.