🔍 Diagnóstico ComfyUI

1. Conectividad básica

✅ ComfyUI responde correctamente
{
    "system": {
        "os": "posix",
        "ram_total": 66844590080,
        "ram_free": 60332630016,
        "comfyui_version": "0.3.43",
        "python_version": "3.12.3 (main, Mar  3 2026, 12:15:18) [GCC 13.3.0]",
        "pytorch_version": "2.7.1+cu128",
        "embedded_python": false,
        "argv": [
            "main.py",
            "--listen",
            "0.0.0.0",
            "--port",
            "8192",
            "--preview-method",
            "auto",
            "--gpu-only",
            "--disable-auto-launch"
        ]
    },
    "devices": [
        {
            "name": "cuda:0 NVIDIA GeForce RTX 5090 : cudaMallocAsync",
            "type": "cuda",
            "index": 0,
            "vram_total": 33679736832,
            "vram_free": 33135591424,
            "torch_vram_total": 0,
            "torch_vram_free": 0
        }
    ]
}

2. Modelos de Upscaling

✅ UpscaleModelLoader disponible
Modelos disponibles:
✅ ImageUpscaleWithModel disponible
✅ ImageResize+ disponible

3. Probar Workflow Simple

Respuesta del servidor (HTTP 400):
{"error": {"type": "prompt_no_outputs", "message": "Prompt has no outputs", "details": "", "extra_info": {}}, "node_errors": {}}

4. Historial Reciente

⚠️ No hay historial disponible

5. Cola de Trabajos

✅ Cola accesible
{
    "queue_running": [],
    "queue_pending": []
}

📋 Resumen de Diagnóstico

Revisa los puntos marcados con ❌ o ⚠️ para identificar el problema.

Si ComfyUI responde pero fallan los workflows, puede ser: