fix: add /v1/props and /v1/models/<id> endpoints

router/router.py

@@ -299,6 +299,38 @@ def chat():
 @app.route("/v1/models")
 def models(): return jsonify({"object":"list","data":[{"id":m,"object":"model","owned_by":"syslog","status":check_gpu_health(m).get("status"),"gpu":check_gpu_health(m).get("gpu_name")} for m in GPU_URLS]})
 
+@app.route("/v1/props")
+def props():
+    """Ollama-compatible model properties endpoint."""
+    props = {}
+    for m in GPU_URLS:
+        h = check_gpu_health(m)
+        props[m] = {
+            "status": h.get("status", "unknown"),
+            "gpu": h.get("gpu_name", m),
+            "max_concurrent": GPU_MAX_CONCURRENT.get(m, 1),
+            "active_requests": gpu_active_count(m),
+            "vram_used_mb": h.get("vram_used_mb"),
+            "vram_total_mb": h.get("vram_total_mb"),
+        }
+    return jsonify({"models": props})
+
+@app.route("/v1/models/<model_id>")
+def model_detail(model_id):
+    """Single model detail — Ollama-compatible."""
+    if model_id in GPU_URLS:
+        h = check_gpu_health(model_id)
+        return jsonify({
+            "id": model_id,
+            "object": "model",
+            "owned_by": "syslog",
+            "status": h.get("status"),
+            "gpu": h.get("gpu_name"),
+            "max_concurrent": GPU_MAX_CONCURRENT.get(model_id, 1),
+            "active_requests": gpu_active_count(model_id),
+        })
+    return jsonify({"error": "model not found"}), 404
+
 @app.route("/health")
 def health():
     gpus = {}