Start a llama-server
start_serverStart a llama-server process for a GGUF model, ensuring the port is free and the server is healthy before proceeding.
Instructions
Spawn llama-server for a model (full arg validation, no shell), wait for its /health endpoint to report ready, and record the pid under bellows supervision. Refuses to start if the port is already in use.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"model": "<model>"
},
"name": "start_server"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| ctx | No | Context window in tokens. | |
| ngl | No | GPU layers to offload (99 = all). | |
| port | No | HTTP port for llama-server. | |
| model | Yes | Model to serve: an absolute path to a .gguf file, or a substring matched against the scanned models list (e.g. 'lfm2.5' or 'Q4_K_M'). | |
| readyTimeoutMs | No | How long to wait for the /health endpoint before giving up and killing the process. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| pid | Yes | ||
| port | Yes | ||
| apiBase | Yes | ||
| modelPath | Yes | ||
| startedAt | Yes |