models.providers.<id>.localService starts a provider-owned local model server on demand. When a model or embedding request selects that provider, Apus checks the health endpoint, starts the process if it is down, waits for readiness, then sends the request. Use it to avoid keeping expensive local servers running all day.

How it works

1. A model or embedding request resolves to a configured provider.
2. If that provider has localService, Apus checks healthUrl.
3. On a successful check, Apus uses the already-running server.
4. On a failed check, Apus spawns command with args.
5. Apus polls the health endpoint until readyTimeoutMs expires.
6. The request goes through the normal model or embedding transport.
7. If Apus started the process and idleStopMs is set, it stops the process after the last in-flight request has been idle that long.

Apus does not install launchd, systemd, Docker, or any daemon for this. The server is a plain child process of whichever Apus process first needed it.

Startup is serialized per configured provider and command/argument/env set, so concurrent chat and embedding requests for the same service do not spawn duplicate servers. Each request holds its own lease until response handling completes, so idle shutdown waits for every in-flight model and embedding request. Configured provider aliases remain distinct: two aliases can point at different GPU hosts without collapsing onto the same Ollama, LM Studio, or OpenAI-compatible adapter id.

Apus waits for any idle shutdown already in progress when it closes local services. A new request for the same service waits for that stop before acquiring a replacement. Shutdown errors remain visible; subsequent requests recheck the owned process before starting a replacement.

Shutdown completion requires the child and its output streams to close and pending tree-termination operations to finish. A missing PID alone does not release the service for replacement.

Final exit observation starts after bounded tree-termination helpers finish. On Windows, this lets taskkill complete without reporting an already-closed service as a shutdown failure.

If another Apus process already has a healthy server at the same healthUrl, this process reuses it without adopting it (each process only manages the child it personally started). Startup and exit logs include bounded, redacted child-output tails plus timing and exit details; configured environment values are never emitted.

For its pinned managed installation, llama.cpp checks for orphaned servers before the first local-service request in a new host process. On native Apple silicon macOS, it reclaims a launchd-adopted router only when the native executable is exactly this state directory's managed binary and its explicit host, port, and preset arguments match; relative presets also require the configured working directory. It captures matching descendants, sends SIGTERM, waits up to five seconds, then uses SIGKILL if needed. Process birth and executable identities are rechecked before each signal. Live parents, custom binaries, other state directories, configurations without unique explicit host/port/preset arguments, and different ports or presets are left alone. A new managed child then starts normally, including after a previous Gateway died from SIGHUP, SIGKILL, or a crash. This does not change inherited nohup signal handling.

Update the host and plugin together to enable recovery; older compatible hosts without the recovery capability keep their existing reuse behavior. Hosts without native process identity support, Linux, and Windows retain the existing reuse behavior; Linux PID 1 may itself be a live Gateway, so parent PID alone cannot identify an orphan. On a supported host, if a matching orphan cannot be safely identified, startup reports its PID and asks you to stop it manually before retrying.

Managed llama.cpp

The official llama.cpp provider generates this shape automatically. Its guided setup installs a pinned, verified llama-server, writes an absolute command and router preset, selects a free loopback port, and stores the resulting baseUrl and localService config. Chat and local embeddings lease the same managed router through the normal OpenAI-compatible transports.

Do not copy a generated command path between machines. Run llama.cpp setup on each Gateway host so Apus selects and verifies the matching platform build. See llama.cpp Provider.

Config shape

{
  models: {
    providers: {
      local: {
        baseUrl: "http://127.0.0.1:8000/v1",
        apiKey: "local-model",
        api: "openai-completions",
        timeoutSeconds: 300,
        localService: {
          command: "/absolute/path/to/server",
          args: ["--host", "127.0.0.1", "--port", "8000"],
          cwd: "/absolute/path/to/working-dir",
          env: { LOCAL_MODEL_CACHE: "/absolute/path/to/cache" },
          healthUrl: "http://127.0.0.1:8000/v1/models",
          readyTimeoutMs: 180000,
          idleStopMs: 0,
        },
        models: [
          {
            id: "my-local-model",
            name: "My Local Model",
            reasoning: false,
            input: ["text"],
            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
            contextWindow: 131072,
            maxTokens: 8192,
          },
        ],
      },
    },
  },
}

Set timeoutSeconds on the provider entry (not localService) so slow cold starts and long generations do not hit the default model request timeout. Set an explicit healthUrl whenever your server exposes readiness somewhere other than /models on the base URL.

During memory_search, managed embedding startup uses readyTimeoutMs instead of the search and query-embedding timeouts. Those timers resume after the service is ready. Embedding requests, retrieval, and result processing keep their existing time limits. Concurrent wiki searches and manager cleanup keep separate limits, and caller cancellation can still stop startup.

Fields

llmman example

llmman is a custom OpenAI-compatible /v1 backend, so the same localService API works with an llmman provider entry. It listens on 127.0.0.1:17434 by default; LLMMAN_HOST overrides the bind address, while LLMMAN_LLM_LIBRARY overrides GPU auto-detection. Its API has no authentication, so keep the default loopback bind unless a trusted network boundary restricts access. llmman has no /health route; use /v1/models or /api/version as healthUrl.

{
  agents: {
    defaults: {
      model: { primary: "llmman/qwen3.8" },
    },
  },
  models: {
    mode: "merge",
    providers: {
      llmman: {
        baseUrl: "http://127.0.0.1:17434/v1",
        apiKey: "${LLMMAN_API_KEY}",
        api: "openai-completions",
        timeoutSeconds: 300,
        localService: {
          command: "/opt/homebrew/bin/llmman",
          args: ["serve"],
          env: { LLMMAN_CONTEXT_LENGTH: "65536" },
          healthUrl: "http://127.0.0.1:17434/v1/models",
          readyTimeoutMs: 180000,
          idleStopMs: 0,
        },
        models: [
          {
            id: "qwen3.8",
            name: "Qwen3.8 (llmman)",
            reasoning: true,
            input: ["text", "image"],
            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
            contextWindow: 65536,
            maxTokens: 8192,
          },
        ],
      },
    },
  },
}

Replace command with the result of which llmman on the machine running Apus, and set LLMMAN_API_KEY=llmman-local in ~/.apus/.env. Bare llmman serve requires no model argument and loads the model on the first request that names it; an optional model argument preloads it instead. Daemon settings such as LLMMAN_CONTEXT_LENGTH go in env. Full llmman setup, including hybrid local + hosted routing: llmman.

ds4 example

{
  models: {
    providers: {
      ds4: {
        baseUrl: "http://127.0.0.1:18000/v1",
        apiKey: "ds4-local",
        api: "openai-completions",
        timeoutSeconds: 300,
        localService: {
          command: "<DS4_DIR>/ds4-server",
          args: [
            "--model",
            "<DS4_DIR>/ds4flash.gguf",
            "--host",
            "127.0.0.1",
            "--port",
            "18000",
            "--ctx",
            "32768",
            "--tokens",
            "128",
          ],
          cwd: "<DS4_DIR>",
          healthUrl: "http://127.0.0.1:18000/v1/models",
          readyTimeoutMs: 300000,
          idleStopMs: 0,
        },
        models: [],
      },
    },
  },
}

Full setup, context sizing, and verification commands: ds4.

Related

Local model setup, provider choices, and safety guidance.

Local models, hybrid local + hosted routing, and on-demand startup with llmman.