[
  {
    "build_commit": "7fe450e",
    "build_number": 1,
    "cpu_info": "AMD Ryzen 7 5800X3D 8-Core Processor",
    "gpu_info": "NVIDIA GeForce RTX 5080",
    "backends": "CUDA",
    "model_filename": "${HOME}/.local/share/llama.cpp/models/Qwen3-8B-Q4_K_M.gguf",
    "model_type": "qwen3 8B Q4_K - Medium",
    "model_size": 5021827072,
    "model_n_params": 8190735360,
    "n_batch": 2048,
    "n_ubatch": 512,
    "n_threads": 8,
    "cpu_mask": "0x0",
    "cpu_strict": false,
    "poll": 50,
    "type_k": "f16",
    "type_v": "f16",
    "n_gpu_layers": 99,
    "n_cpu_moe": 0,
    "split_mode": "layer",
    "main_gpu": 0,
    "no_kv_offload": false,
    "flash_attn": 1,
    "devices": "CUDA0",
    "tensor_split": "0.00",
    "tensor_buft_overrides": "none",
    "load_mode": "auto",
    "lazy_mode": "auto",
    "embeddings": false,
    "no_op_offload": 0,
    "no_host": false,
    "fit_target": 0,
    "fit_min_ctx": 0,
    "n_prompt": 2048,
    "n_gen": 0,
    "n_depth": 0,
    "test_time": "2026-09-26T20:57:58Z",
    "avg_ns": 240516468,
    "stddev_ns": 2865645,
    "avg_ts": 8515.967319,
    "stddev_ts": 100.503444,
    "samples_ns": [
      245229887,
      239264372,
      240507113,
      240036196,
      237544776
    ],
    "samples_ts": [
      8351.35,
      8559.57,
      8515.34,
      8532.05,
      8621.53
    ]
  },
  {
    "build_commit": "7fe450e",
    "build_number": 1,
    "cpu_info": "AMD Ryzen 7 5800X3D 8-Core Processor",
    "gpu_info": "NVIDIA GeForce RTX 5080",
    "backends": "CUDA",
    "model_filename": "${HOME}/.local/share/llama.cpp/models/Qwen3-8B-Q4_K_M.gguf",
    "model_type": "qwen3 8B Q4_K - Medium",
    "model_size": 5021827072,
    "model_n_params": 8190735360,
    "n_batch": 2048,
    "n_ubatch": 512,
    "n_threads": 8,
    "cpu_mask": "0x0",
    "cpu_strict": false,
    "poll": 50,
    "type_k": "f16",
    "type_v": "f16",
    "n_gpu_layers": 99,
    "n_cpu_moe": 0,
    "split_mode": "layer",
    "main_gpu": 0,
    "no_kv_offload": false,
    "flash_attn": 1,
    "devices": "CUDA0",
    "tensor_split": "0.00",
    "tensor_buft_overrides": "none",
    "load_mode": "auto",
    "lazy_mode": "auto",
    "embeddings": false,
    "no_op_offload": 0,
    "no_host": false,
    "fit_target": 0,
    "fit_min_ctx": 0,
    "n_prompt": 0,
    "n_gen": 256,
    "n_depth": 0,
    "test_time": "2026-09-26T20:58:01Z",
    "avg_ns": 1450187828,
    "stddev_ns": 656228,
    "avg_ts": 176.528886,
    "stddev_ts": 0.079807,
    "samples_ns": [
      1451085357,
      1449689508,
      1450197894,
      1449443790,
      1450522592
    ],
    "samples_ts": [
      176.42,
      176.59,
      176.528,
      176.619,
      176.488
    ]
  },
  {
    "build_commit": "7fe450e",
    "build_number": 1,
    "cpu_info": "AMD Ryzen 7 5800X3D 8-Core Processor",
    "gpu_info": "NVIDIA GeForce RTX 5080",
    "backends": "CUDA",
    "model_filename": "${HOME}/.local/share/llama.cpp/models/Qwen3-8B-Q4_K_M.gguf",
    "model_type": "qwen3 8B Q4_K - Medium",
    "model_size": 5021827072,
    "model_n_params": 8190735360,
    "n_batch": 2048,
    "n_ubatch": 512,
    "n_threads": 8,
    "cpu_mask": "0x0",
    "cpu_strict": false,
    "poll": 50,
    "type_k": "f16",
    "type_v": "f16",
    "n_gpu_layers": 99,
    "n_cpu_moe": 0,
    "split_mode": "layer",
    "main_gpu": 0,
    "no_kv_offload": false,
    "flash_attn": 1,
    "devices": "CUDA0",
    "tensor_split": "0.00",
    "tensor_buft_overrides": "none",
    "load_mode": "auto",
    "lazy_mode": "auto",
    "embeddings": false,
    "no_op_offload": 0,
    "no_host": false,
    "fit_target": 0,
    "fit_min_ctx": 0,
    "n_prompt": 2048,
    "n_gen": 0,
    "n_depth": 4096,
    "test_time": "2026-09-26T20:58:08Z",
    "avg_ns": 279111710,
    "stddev_ns": 2524525,
    "avg_ts": 7338.043574,
    "stddev_ts": 66.295577,
    "samples_ns": [
      276123073,
      281249964,
      278302896,
      282172697,
      277709920
    ],
    "samples_ts": [
      7416.98,
      7281.78,
      7358.89,
      7257.97,
      7374.6
    ]
  },
  {
    "build_commit": "7fe450e",
    "build_number": 1,
    "cpu_info": "AMD Ryzen 7 5800X3D 8-Core Processor",
    "gpu_info": "NVIDIA GeForce RTX 5080",
    "backends": "CUDA",
    "model_filename": "${HOME}/.local/share/llama.cpp/models/Qwen3-8B-Q4_K_M.gguf",
    "model_type": "qwen3 8B Q4_K - Medium",
    "model_size": 5021827072,
    "model_n_params": 8190735360,
    "n_batch": 2048,
    "n_ubatch": 512,
    "n_threads": 8,
    "cpu_mask": "0x0",
    "cpu_strict": false,
    "poll": 50,
    "type_k": "f16",
    "type_v": "f16",
    "n_gpu_layers": 99,
    "n_cpu_moe": 0,
    "split_mode": "layer",
    "main_gpu": 0,
    "no_kv_offload": false,
    "flash_attn": 1,
    "devices": "CUDA0",
    "tensor_split": "0.00",
    "tensor_buft_overrides": "none",
    "load_mode": "auto",
    "lazy_mode": "auto",
    "embeddings": false,
    "no_op_offload": 0,
    "no_host": false,
    "fit_target": 0,
    "fit_min_ctx": 0,
    "n_prompt": 0,
    "n_gen": 256,
    "n_depth": 4096,
    "test_time": "2026-09-26T20:58:11Z",
    "avg_ns": 1627970135,
    "stddev_ns": 1690501,
    "avg_ts": 157.251177,
    "stddev_ts": 0.162999,
    "samples_ns": [
      1630919211,
      1627547497,
      1627117690,
      1626677784,
      1627588497
    ],
    "samples_ts": [
      156.967,
      157.292,
      157.333,
      157.376,
      157.288
    ]
  }
]