TTS-Slots (Piper CPU, Qwen3-TTS auf Ada), M4000 außer Betrieb, Engine-Grenzen im Placement (ADR-0006)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 02:12:30 +02:00
parent 0533f224f1
commit f90b7df31c
25 changed files with 586 additions and 23 deletions
@@ -25,7 +25,7 @@ class SlotSpec(BaseModel):
    effective_context: int | None = None
    supports_lora: bool = False
    fallbacks: list[str] = Field(default_factory=list)
-    kind: Literal["chat", "embedding", "rerank"] = "chat"
+    kind: Literal["chat", "embedding", "rerank", "tts"] = "chat"
    canary: bool = True            # vom Prober geprüft (für tier 0 standardmäßig aus, siehe canary.py)
    ops: dict[str, str] = Field(default_factory=dict)  # docker_host, container, pve_node, vmid → Healer-Labels