[
  {
    "build_commit": "7fe450e19",
    "build_number": 11146,
    "cpu_info": "Accelerate, Apple M1 Pro",
    "gpu_info": "Apple M1 Pro",
    "backends": "BLAS,MTL",
    "model_filename": "<local-path-redacted>/sha256-6a0746a1ec1aef3e7ec53868f220ff6e389f6f8ef87a01d77c96807de94ca2aa",
    "model_type": "llama 8B Q4_0",
    "model_size": 4653375488,
    "model_n_params": 8030261248,
    "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": -1,
    "n_cpu_moe": 0,
    "split_mode": "layer",
    "main_gpu": 0,
    "no_kv_offload": false,
    "flash_attn": -1,
    "devices": "auto",
    "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": 512,
    "n_gen": 0,
    "n_depth": 8192,
    "test_time": "2026-09-27T16:59:18Z",
    "avg_ns": 2689215283,
    "stddev_ns": 3884331,
    "avg_ts": 190.390430,
    "stddev_ts": 0.274684,
    "samples_ns": [ 2695439708, 2685869166, 2688734834, 2686106417, 2689926292 ],
    "samples_ts": [ 189.95, 190.627, 190.424, 190.61, 190.34 ]
  },
  {
    "build_commit": "7fe450e19",
    "build_number": 11146,
    "cpu_info": "Accelerate, Apple M1 Pro",
    "gpu_info": "Apple M1 Pro",
    "backends": "BLAS,MTL",
    "model_filename": "<local-path-redacted>/sha256-6a0746a1ec1aef3e7ec53868f220ff6e389f6f8ef87a01d77c96807de94ca2aa",
    "model_type": "llama 8B Q4_0",
    "model_size": 4653375488,
    "model_n_params": 8030261248,
    "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": -1,
    "n_cpu_moe": 0,
    "split_mode": "layer",
    "main_gpu": 0,
    "no_kv_offload": false,
    "flash_attn": -1,
    "devices": "auto",
    "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": 128,
    "n_depth": 8192,
    "test_time": "2026-09-27T17:00:08Z",
    "avg_ns": 4585603666,
    "stddev_ns": 38634094,
    "avg_ts": 27.915013,
    "stddev_ts": 0.232568,
    "samples_ns": [ 4654656833, 4566227583, 4568949083, 4570498667, 4567686167 ],
    "samples_ts": [ 27.4993, 28.0319, 28.0152, 28.0057, 28.0229 ]
  }
]