# Wicklee (wicklee.dev)

Measures watts and tokens per node to show true cost per model and idle hardware waste for private AI fleets.

- Atlas card: https://inite.ai/en/atlas/wicklee.dev
- Site: https://wicklee.dev
- For: Teams running private AI model fleets on their own hardware who need node health and cost visibility.
- Use 18/100 · alive 70/100

## Used for

- find idle hardware drawing power while no model runs
- pick the healthiest node before sending a request
- measure real cost per model from watts and throughput

## Price

Not read yet.

## For developers

- OpenAPI: https://wicklee.dev/openapi.json (19 paths)
- MCP server: https://wicklee.dev/mcp (not in the official registry; auth on 2026-10-28)
- Read 2026-10-22

## MCP server, measured daily

Server: https://wicklee.dev/mcp, checked every day since 2026-10-28 (7 checks).

- Answered: 100% of daily checks
- Median first answer: 242 ms
- Sign-in: unknown
- Method: https://inite.ai/en/atlas/mcp/report

## Your data

- Certifications stated: none
- Read 2026-10-22

---

Source: INITE Atlas, https://inite.ai/en/atlas/wicklee.dev. JSON: https://inite.ai/api/atlas/wicklee.dev. MCP: https://inite.ai/api/mcp (atlas_lookup).