feat: display LLM metrics and capabilities (Feature 6)

routers/models.py

@@ -16,6 +16,8 @@ from auth import auth
 from config import CMD_TEMPLATE, CONFIG_PATH, DEFAULT_TTL, LLAMA_SWAP_URL, MODELS_DIR
 from jobengine import JOBS, start_job
 from llamaswap import _swap_get, read_config, write_config
+import re
+import os
 
 router = APIRouter(prefix="/api", dependencies=[Depends(auth)])
 
@@ -50,12 +52,42 @@ def status():
     configured = {}
     for name, spec in (cfg.get("models") or {}).items():
         spec = spec or {}
+        cmd = str(spec.get("cmd", "")).strip()
+        
+        # Parse Meta
+        ctx = 8192
+        m_ctx = re.search(r'-(?:c|-ctx-size)\s+(\d+)', cmd)
+        if m_ctx: ctx = int(m_ctx.group(1))
+        
+        size_bytes = None
+        quant = ""
+        m_path = re.search(r'-(?:m|-model)\s+([^\s]+)', cmd)
+        if m_path:
+            path = m_path.group(1).replace("'", "").replace('"', '')
+            if os.path.exists(path):
+                size_bytes = os.path.getsize(path)
+            q_match = re.search(r'(Q\d_[A-Z0-9_]+|IQ\d_[A-Z0-9_]+|fp16|bf16)\.gguf', path, flags=re.IGNORECASE)
+            if q_match:
+                quant = q_match.group(1).upper()
+                
+        caps = ["Text"]
+        if "coder" in name.lower() or (m_path and "code" in m_path.group(1).lower()):
+            caps = ["Code"]
+        if "--mmproj" in cmd:
+            caps.append("Bild")
+            
         configured[name] = {
             "name": name,
             "ttl": spec.get("ttl", cfg.get("globalTTL", 0)),
-            "cmd": str(spec.get("cmd", "")).strip(),
+            "cmd": cmd,
             "state": "idle",
             "port": None,
+            "meta": {
+                "ctx": ctx,
+                "size_bytes": size_bytes,
+                "quant": quant,
+                "caps": caps
+            }
         }
     swap_ok = True
     try: