> ## Documentation Index
> Fetch the complete documentation index at: https://docs.almond.bot/llms.txt
> Use this file to discover all available pages before exploring further.

# inference-server

> Serve policy inference on a separate, more powerful machine for run-policy.

Runs LeRobot's async-inference `PolicyServer` in the foreground, so [`run-policy`](/cli/run-policy) can offload inference to a more powerful machine (e.g. a desktop with a discrete GPU) on the same network as the robot. The robot streams joint positions + camera frames to it over gRPC and receives action chunks back.

The policy itself (`--policy_path` / `--policy_type` / `--device`) is selected by the **client** (`run-policy`), not the server — so one server can serve different policies across sessions without restarting. The server downloads the policy itself, so the path must be reachable from it (e.g. a HuggingFace Hub repo ID, or a local path that exists on the server machine).

```bash theme={null}
axol inference-server                 # listen on 0.0.0.0:8765
axol inference-server --port 9000
```

Then, on the robot:

```bash theme={null}
axol run-policy --server_host <server-ip> --policy_path myorg/pick-place-policy \
    --policy_type act --task "Pick the red cube" ...
```

| Flag                                     | Description                                                                                           |
| ---------------------------------------- | ----------------------------------------------------------------------------------------------------- |
| `--host HOST`                            | Interface to bind the gRPC server to. Default `0.0.0.0` (accepts connections from the whole network). |
| `--port INT`                             | gRPC port (default: 8765). Must match `run-policy`'s `--server_port`.                                 |
| `--fps INT`                              | Action chunk rate (default: 60). Must match `run-policy`'s `--fps`.                                   |
| `--log_level {DEBUG,INFO,WARNING,ERROR}` | Default: `INFO`                                                                                       |

<Note>
  Requires the `lerobot` extra and, for GPU inference, a CUDA-capable PyTorch install on the server machine. Without `--server_host`, `run-policy` spawns its own local `PolicyServer` child process — this command is only needed when inference runs on a different machine.
</Note>