{
  "object": "list",
  "data": [
    {
      "object": "model_capacity",
      "id": "primary-local",
      "aliases": [
        "llm.auxiliary",
        "llm.primary",
        "llm.voice",
        "vision.general",
        "vision.ocr"
      ],
      "model": "glm53-flash-exl3-4bpw-v84-fp8-tp2-dcp1-c4-327k-r11-vision8-apc-nospec",
      "loaded": true,
      "readiness": {
        "state": "ready",
        "reason": "identity_passed"
      },
      "engine": {
        "name": "vllm-glm53-exl3-b12x-v84-fp8-control",
        "quantization": "exl3-4bpw-fp8-ds-mla-kv"
      },
      "gpu": {
        "role": null,
        "name": null,
        "memory_total_mib": 195774
      },
      "capacity": {
        "context_limit_tokens": 327680,
        "kv_cache_capacity_tokens": null,
        "full_context_concurrency": null,
        "configured_max_concurrency": 4,
        "scheduler_max_num_seqs": 4,
        "model_memory_gib": null
      },
      "multimodal": {
        "admission_enabled": true,
        "image_limit": 8,
        "video_limit": 0,
        "image_tokens_estimate": 8192,
        "video_tokens_estimate": 0
      },
      "live": {
        "status": "available",
        "reason": null,
        "requests_running": 1.0,
        "requests_waiting": 0.0,
        "kv_cache_usage_fraction": 0.022140221402214055,
        "kv_cache_used_tokens_estimate": null,
        "kv_cache_remaining_tokens_estimate": null,
        "preemptions_total": 0.0,
        "multimodal_cache_queries_total": 48.0,
        "multimodal_cache_hits_total": 46.0,
        "generation_tokens_total": 372990.0,
        "generation_throughput_tokens_per_second": null,
        "prefix_cache_queries_total": 58777875.0,
        "prefix_cache_hits_total": 56446208.0,
        "prefix_cache_hit_rate": null
      },
      "scenario": null
    },
    {
      "object": "model_capacity",
      "id": "secondary-node",
      "aliases": [
        "llm.secondary"
      ],
      "model": "thinkingcap-awq-5090",
      "loaded": true,
      "readiness": {
        "state": "ready",
        "reason": "identity_passed"
      },
      "engine": {
        "name": "vllm",
        "quantization": "nvfp4-awq-w4a4-bf16-vision-mtp3-bf16-kv-uva3"
      },
      "gpu": {
        "role": "secondary-compute",
        "name": "NVIDIA GeForce RTX 5090",
        "memory_total_mib": 32607
      },
      "capacity": {
        "context_limit_tokens": 131072,
        "kv_cache_capacity_tokens": 138519,
        "full_context_concurrency": 1.057,
        "configured_max_concurrency": 1,
        "scheduler_max_num_seqs": 1,
        "model_memory_gib": null
      },
      "multimodal": {
        "admission_enabled": true,
        "image_limit": 2,
        "video_limit": 0,
        "image_tokens_estimate": 3072,
        "video_tokens_estimate": 0
      },
      "live": {
        "status": "available",
        "reason": null,
        "requests_running": 0.0,
        "requests_waiting": 0.0,
        "kv_cache_usage_fraction": 0.0,
        "kv_cache_used_tokens_estimate": 0,
        "kv_cache_remaining_tokens_estimate": 138519,
        "preemptions_total": 0.0,
        "multimodal_cache_queries_total": 30.0,
        "multimodal_cache_hits_total": 23.0,
        "generation_tokens_total": 58586.0,
        "generation_throughput_tokens_per_second": null,
        "prefix_cache_queries_total": 0.0,
        "prefix_cache_hits_total": 0.0,
        "prefix_cache_hit_rate": null
      },
      "scenario": null
    }
  ]
}
