# Microgram: cognition-cognition-perplexity

`station__algovigilance__microgram-cognition-cognition-perplexity` · native (always callable) · domain `algovigilance` · pv-relevance `pv-core`

/// Perplexity = exp(-1/N · Σ log(p(tokenᵢ)))

## Agent metadata

- `idempotent`: true
- `read_only`: true
- `expected_latency_ms`: unknown (not yet contract-tested)
- `cost_tokens_estimate`: unknown

## Input schema

- `logits_per_step` *object* (required) — Logit vectors for each step (each inner vec has vocab_size elements).
- `token_ids` *object* (required) — Token IDs of the sequence.

## Example call

```json
POST /api/mcp
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "station__algovigilance__microgram-cognition-cognition-perplexity",
    "arguments": {
      "logits_per_step": {},
      "token_ids": {}
    }
  }
}
```

## Related

- [/tools](/tools) — all 7718 tools
- [/tools/algovigilance__microgram-cognition-cognition-perplexity](/tools/algovigilance__microgram-cognition-cognition-perplexity) — HTML page
- [/tools/algovigilance__microgram-cognition-cognition-perplexity/json](/tools/algovigilance__microgram-cognition-cognition-perplexity/json) — JSON form (agent-friendly)
- [/api/mcp](/api/mcp) — endpoint
- [/AGENTS.md](/AGENTS.md) — agent guide