LlmCore.LLM.Appliance (llm_core v0.3.0)

Generic local inference appliance provider (DGX Spark, future devices).

Assumes an OpenAI-compatible chat completion API exposed over HTTP.

Summary

Functions

available?()

Checks if the appliance base URL is configured and the health endpoint responds.

capabilities()

Returns the appliance capability map. Models and max_context are sourced from application config.

discover()

Discover configured appliances (future mDNS hook).

health(url)

Perform a lightweight health check against the appliance.

provider_type()

Returns :local — Appliance is a local inference provider.

send(prompt, opts \\ [])

Sends a prompt to the appliance chat completions endpoint.

stream(prompt, opts \\ [])

Streams a response from the appliance chat completions endpoint via SSE.

Functions

available?()

@spec available?() :: boolean()

Checks if the appliance base URL is configured and the health endpoint responds.

capabilities()

@spec capabilities() :: LlmCore.LLM.Provider.capabilities()

Returns the appliance capability map. Models and max_context are sourced from application config.

discover()

@spec discover() :: [{String.t() | binary(), URI.t()}]

Discover configured appliances (future mDNS hook).

health(url)

@spec health(String.t() | URI.t()) :: boolean()

Perform a lightweight health check against the appliance.

provider_type()

@spec provider_type() :: :local

Returns :local — Appliance is a local inference provider.

send(prompt, opts \\ [])

@spec send(
  LlmCore.LLM.Provider.prompt(),
  keyword()
) :: {:ok, LlmCore.LLM.Response.t()} | {:error, LlmCore.LLM.Error.t()}

Sends a prompt to the appliance chat completions endpoint.

stream(prompt, opts \\ [])

@spec stream(
  LlmCore.LLM.Provider.prompt(),
  keyword()
) :: {:ok, Enumerable.t()} | {:error, LlmCore.LLM.Error.t()}

Streams a response from the appliance chat completions endpoint via SSE.