Interface: ModelConfig
Defined in: packages/core/src/providers/types.ts:167
Model configuration with provider details
Examples
Simple string (uses env vars)
model: 'gpt-5.6-luna'
With provider config
model: {
id: 'gpt-5.6-luna',
provider: 'openai',
config: {
apiKey: 'sk-...',
fetch: customFetch
}
}
Properties
cacheControl?
optionalcacheControl?:CacheControlOption
Defined in: packages/core/src/providers/types.ts:219
Construction-time prompt-cache plan. true is the chat shorthand
(last-tool + system + last four, then capped). An object is an
explicit plan — { scope: 'prefix' } stamps last-tool + system.
false turns it off. Per-call cache / cacheControl /
enableCaching override this.
config?
optionalconfig?:ProviderConfig
Defined in: packages/core/src/providers/types.ts:177
id
id:
string
Defined in: packages/core/src/providers/types.ts:168
provider?
optionalprovider?:ProviderName
Defined in: packages/core/src/providers/types.ts:175
Provider name. When omitted it is read from the id: an explicit
provider|id / provider:id prefix, or a vendor-owned id prefix from the
catalog (gpt- → openai, claude- → anthropic, …). Other ids need it.
reasoning?
optionalreasoning?:object
Defined in: packages/core/src/providers/types.ts:194
Written into each provider's own form by its catalog reasoning
capability; providers without one ignore it.
budgetTokens?
optionalbudgetTokens?:number
Token budget for thinking. Wins over effort where the provider takes a
budget; ignored by Claude models that think adaptively (4.6+).
effort?
optionaleffort?:"max"|"low"|"medium"|"high"|"xhigh"|string&object
The named levels are the common ones and are what autocomplete offers,
but the set is the provider's, not ours: Anthropic advertises xhigh
and max for claude-opus-5, and publishes the list per model under
capabilities.effort — which listModels reports as ModelInfo.effort.
A closed union here would reject levels the provider documents, so the
type suggests without forbidding.
summary?
optionalsummary?:"concise"|"detailed"|"auto"
Example
model: {
id: 'gpt-5',
reasoning: {
effort: 'high',
summary: true
}
}