{"service":"wave Dispatch","role":"routing decision only — control plane","data_policy":{"sees":"only the first ~2000 chars of your prompt (≈512 tokens, the bge-base-en routing model's limit) — used ONLY to pick a model. This is a routing PEEK, not a cap on your prompt: your FULL prompt runs on YOUR infra / frontier, never limited or stored by us.","logs":["per-license daily request COUNT (use:<key>:<YYYYMMDD>)","license metadata (plan/limit/email)"],"never_logs":["prompt content","your data","your API keys","model outputs"],"data_plane":"your API keys, inference, and data stay on YOUR infra. We return only {route, probability, margin}.","sovereign":"with a Sovereign profile, the local→heavy→frontier chain runs entirely on YOUR endpoints; the edge adds only {profile, chain, params} to the decision. Send a pre-computed vector to skip even the routing peek — then NOTHING leaves your network."},"data_use":{"default":"PRIVATE for everyone (all plans). Your prompt routes a request in-the-moment; it is NOT stored or used to train any central/shared model.","self_improvement":"happens LOCALLY on your own instance (your traffic improves your router); your data never leaves your infra.","opt_in":"you may OPT IN to contribute anonymized routing labels (the route decision, not raw prompts) to improve the shared router — for account credit. Never opt-out, never default."},"edge_model":"@cf/baai/bge-base-en-v1.5 (embedding only, at the edge)","source":"open edge worker + SDKs, auditable; logs zero prompt content (most requests never leave your infra)","version":"see deploy"}