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

# Targets

> List API Performance text models, coding agents, and target objects from the Lightsage CLI.

Targets are the runnable model choices available to API Performance evals. Use them to discover what you can pass to `lightsage api-performance config update --targets`.

Prompt tracking uses answer-engine platforms from `lightsage prompts config retrieve`. API Performance uses targets because coding agents have both an execution platform and an underlying model.

## Command

| Command                                  | Description                          |
| ---------------------------------------- | ------------------------------------ |
| `lightsage api-performance targets list` | List API Performance target options. |

## List targets

```bash theme={null}
lightsage api-performance targets list --output-format json
```

Example response:

```json theme={null}
{
  "text_models": [
    {
      "id": "openai/gpt-5-mini",
      "label": "GPT-5 Mini",
      "platform": "openai",
      "cost_per_1m_input": 0.25,
      "cost_per_1m_output": 2.0,
      "target": {
        "target_type": "text",
        "execution_platform": "openrouter",
        "model_id": "openai/gpt-5-mini"
      }
    }
  ],
  "coding_agents": [
    {
      "id": "codex",
      "label": "Codex CLI",
      "platform": "codex",
      "cost_estimate_per_eval": 0.03,
      "default_model_id": "gpt-5.4",
      "target": {
        "target_type": "agent",
        "execution_platform": "codex",
        "model_id": "gpt-5.4"
      },
      "models": [
        {
          "id": "gpt-5.4",
          "label": "GPT-5.4",
          "target": {
            "target_type": "agent",
            "execution_platform": "codex",
            "model_id": "gpt-5.4"
          }
        }
      ]
    }
  ],
  "frequencies": ["daily", "every_other_day", "weekly", "manual"],
  "meta": {
    "text_model_count": 1,
    "coding_agent_count": 1
  }
}
```

## Use a target

Copy a `target` object into config update and add scheduling fields:

```bash theme={null}
lightsage api-performance config update \
  --targets '[{"target_type":"agent","execution_platform":"codex","model_id":"gpt-5.4","frequency":"daily","enabled":true}]'
```

Use `frequency` values from the response. Set `frequency` to `manual` when you want the target available without scheduling it.

## Target types

| Type         | How to configure                                                                                                              |
| ------------ | ----------------------------------------------------------------------------------------------------------------------------- |
| Text model   | Use `target_type` `text`, `execution_platform` `openrouter`, and a `model_id` from `text_models`.                             |
| Coding agent | Use `target_type` `agent`, an `execution_platform` from `coding_agents.id`, and a `model_id` from that agent's `models` list. |