759 B
759 B
03 — Prefill/decode wire protocol
Status: ready-for-agent
Goal
Define and implement the activation protocol needed by both PyTorch and future GGUF backends.
Scope
- Add route/session lifecycle headers.
- Separate
prefillfromdecode-step. - Keep binary bfloat16 activation bodies.
- Preserve relay compatibility.
- Add route id and model artifact hash validation.
Draft Endpoints
POST /sessions/{session_id}/prefillPOST /sessions/{session_id}/decode-stepDELETE /sessions/{session_id}GET /sessions/{session_id}/status
Acceptance
- Old
/forwardremains temporarily or fails with clear version message. - Tests cover relay preservation of session headers.
- Decode-step payload is independent of prompt length.