LanguageModelFactory

Methods

capabilities

async capabilities(options) =>LanguageModelCapabilities

Get the capabilities of the on-device language model.

Options (optional)
model optional string The id of the model to use, undefined to use the default

Returns a new capabilities object

---

create

async create(options) => LanguageModel

Creates a new language model session

Options (optional)
model optional string
gpuEngine: optional AICapabilityGpuEngine
systemPrompt optional string
initialPrompts optional InitialPrompt[] An array of prompts each in the format [ { content: "The prompt content", role: "system" }, { content: "The prompt content", role: "user" }, { content: "The prompt content", role: "assistant" } ]
topK optional number
topP optional number
temperature optional number
repeatPenality optional number
flashAttention optional boolean
contextSize optional number
maxTokens optional number
grammar optional any
signal optional AbortSignal
monitor optional AICreateMonitor

Returns a new LanguageModel session that can be prompted with the pre-provided configuration