Implement prompt caching, speculative decoding config, parallel slots, and dynamic pricing metrics
This commit is contained in:
@@ -86,6 +86,9 @@ export interface ModelInfo {
|
||||
quant: string
|
||||
size_bytes: number | null
|
||||
incomplete: boolean
|
||||
prompt_cache: boolean
|
||||
spec_draft_model: string | null
|
||||
parallel_slots: number
|
||||
capabilities: Capabilities
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user