System Prompts - RunAnywhere Documentation

Through options

val result = RunAnywhere.llm.generate(
    "How do I reverse a string?",
    LlmOptions(systemPrompt = "You are a Kotlin expert. Answer with code, then one sentence.")
)

Through a message

A ChatRole.System message becomes options.systemPrompt automatically:

val messages = listOf(
    ChatMessage(ChatRole.System, "You are a Kotlin expert."),
    ChatMessage(ChatRole.User, "How do I reverse a string?"),
)

val result = RunAnywhere.llm.generate(messages)

The two are equivalent. When both are present, the explicit options.systemPrompt wins.

Writing one that works

Small on-device models follow short, concrete instructions far better than long ones.

// Works
"You are a terse assistant. Answer in one sentence. No preamble."

// Works less well on a 0.6B model
"You are an extremely helpful, knowledgeable, and friendly AI assistant who always strives
to provide the most comprehensive and detailed responses possible…"

Naming the exact output shape usually holds:

LlmOptions(
    systemPrompt = "You are a classifier. Reply with exactly one word: positive, negative, or neutral.",
    temperature = 0.1f
)

For a guaranteed shape rather than a requested one, use structured output.

Reuse

object Prompts {
    val summarizer = LlmOptions(
        systemPrompt = "Summarize in three bullet points. No introduction.",
        maxOutputTokens = 150,
        temperature = 0.3f
    )

    val coder = LlmOptions(
        systemPrompt = "You are a Kotlin expert. Answer with code first.",
        temperature = 0.2f
    )
}

val summary = RunAnywhere.llm.generate(article, Prompts.summarizer)