{"service":"modelforge-llmops","mode":"demo","thesis":"Agents decide what to do; ModelForge decides which weights, where they run, and how we prove it.","spine_role":"Model Plane flagship (ADR-034) — peer to AegisAI / VAP / Enterprise RAG / Content Factory","components":[{"id":"api","label":"ModelForge API","status":"ready","detail":"Vercel /api health + posture + receipts"},{"id":"peft","label":"PEFT / DomainForge","status":"ready","detail":"CUDA PEFT receipt published (see peft_gpu.json honesty — micro LoRA on T4 or DomainForge 7B ladder)."},{"id":"vllm_cuda","label":"CUDA vLLM serve","status":"ready","detail":"TTFT / tok/s from real vLLM (not Architecture Lab Path B)."},{"id":"slm_bakeoff","label":"SLM bake-off","status":"ready","detail":"Published golden-suite bake-off memo."},{"id":"gateway","label":"LLM gateway bridge","status":"ready","detail":"Synthetic RoutingDecision sample (ADR-028/029) — not a production audit export."}],"non_goals":["Foundation-model pretraining from scratch","Claiming vLLM Architecture Lab Path B as CUDA production","Always-on free-tier GPU","Treating peft_smoke.json as a GPU receipt"]}