Gabriel Cosi

Software Engineer

Systems Operational

Running LLMs at home

Building an on-demand inference box.

LLM Usage

Tokens (7d)11,207,823 tokens

Top Model (7d)qwen3.6-35b-a3b

Cluster Metrics

CPU Load34.7%

Memory45.9%

Active Pods201

Uptime37d

SLA (7d)99.87%

Global P99 (7d)525 ms