feat(2.0): Phase 6a — Modell-Install (Download-Jobs + Split + UI)

jobengine portiert (Live-Log + %-Fortschritt aus .incomplete), services/hf.py
(resolve_gguf inkl. Split -of-, Groessen), POST /api/models/install (hf
download als Job + sofortige Registrierung, Split-fest), GET /api/jobs +
cancel. huggingface_hub in requirements. Frontend: Install-Buttons auf
Discover-Karten + Live-Download-Fortschrittsbalken (JobsBar).

Verifiziert: resolve_gguf gegen echte Repos (Qwen3.6-35B-A3B 23GB single,
Qwen3.5-122B-A10B 77GB 3-part split) + Frontend-Build.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

backend/routers/models.py

@@ -4,8 +4,9 @@ import psutil
 from fastapi import APIRouter, HTTPException
 from pydantic import BaseModel
 
-from services import discover, llamaswap
-from services.fit import evaluate_fit, max_ctx_for
+from config import HF_DOWNLOAD_ENV, MODELS_DIR
+from services import discover, hf, jobengine, llamaswap
+from services.fit import evaluate_fit, extract_params_b, max_ctx_for
 
 router = APIRouter(prefix="/api")
 
@@ -60,6 +61,69 @@ def register(req: RegisterReq) -> dict:
     return {"ok": True, "model_id": model_id}
 
 
+class InstallReq(BaseModel):
+    repo: str
+    role: str | None = None
+    quant: str = "Q4_K_M"
+    ctx: int | None = None
+    jinja: bool = False
+    hf_token: str | None = None
+
+
+@router.post("/models/install")
+def install(req: InstallReq) -> dict:
+    """Lädt ein Modell von HuggingFace (Hintergrund-Job) UND trägt es sofort in
+    llama-swap ein (cmd + Rolle-Alias). llama-swap (-watch-config) lädt es, sobald
+    die Datei da ist. Split-GGUFs werden komplett geladen, registriert wird der
+    erste Teil (-00001-of-…)."""
+    info = hf.resolve_gguf(req.repo, req.quant)
+    if not info["first"]:
+        raise HTTPException(404, f"Keine GGUF-Datei für Quant '{req.quant}' in {req.repo} gefunden.")
+
+    subdir = req.repo.split("/")[-1]
+    target = MODELS_DIR / subdir
+    target.mkdir(parents=True, exist_ok=True)
+    model_path = str(target / info["first"])
+    mmproj_path = str(target / info["mmproj"]) if info["mmproj"] else None
+
+    ctx = req.ctx
+    if ctx is None:
+        ram = _ram_gb()
+        ctx = max_ctx_for(extract_params_b(req.repo), req.quant, ram)
+
+    # Sofort registrieren (Datei kommt gleich) — robust gegen -watch-config.
+    try:
+        model_id = llamaswap.register_model(
+            model_path, role=req.role, ctx=ctx, mmproj_path=mmproj_path, jinja=req.jinja)
+    except PermissionError as exc:
+        raise HTTPException(500, str(exc))
+
+    # Download-Job: alle GGUF-Teile (+ mmproj) per --include holen.
+    args = [hf.hf_bin(), "download", req.repo]
+    for f in info["files"]:
+        args.append(f)
+    if info["mmproj"]:
+        args.append(info["mmproj"])
+    args += ["--local-dir", str(target)]
+    env = dict(HF_DOWNLOAD_ENV)
+    if req.hf_token:
+        env["HF_TOKEN"] = req.hf_token
+    job_id = jobengine.start_job(args, f"download {req.repo}", env=env)
+    jobengine.attach_download_progress(job_id, str(target), info["total_bytes"])
+    return {"ok": True, "job_id": job_id, "model_id": model_id, "model_path": model_path,
+            "total_bytes": info["total_bytes"], "files": len(info["files"])}
+
+
+@router.get("/jobs")
+def jobs() -> dict:
+    return {"jobs": jobengine.public_jobs()}
+
+
+@router.post("/jobs/{job_id}/cancel")
+def cancel(job_id: str) -> dict:
+    return {"ok": jobengine.cancel_job(job_id)}
+
+
 @router.get("/groups")
 def groups() -> dict:
     return {"groups": llamaswap.list_groups()}