Upload microsoft_phi-4_2025-05-28-14-09-35.json
Browse filesik llama run with ./llama-server -m ./phi-4-Q8_0.gguf -c 16384 -ngl 99 -np 16 --cont-batching --host 0.0.0.0 --port 8000
benchmarked with sudo docker run --network host -e HF_TOKEN=$HF_TOKEN -v ~/inference-benchmarker-results:/opt/inference-benchmarker/results inference_benchmarker inference-benchmarker --no-console --prompt-options "num_tokens=200,max_tokens=220,min_tokens=180,variance=10" --decode-options "num_tokens=200,max_tokens=220,min_tokens=180,variance=10" --url $URL --rates 1.0 --rates 10.0 --rates 30.0 --rates 100.0 --max-vus 8 --duration 120s --warmup 30s --benchmark-kind rate --model-name "phi-4" --tokenizer-name "microsoft/phi-4"
results/microsoft_phi-4_2025-05-28-14-09-35.json
ADDED
|
@@ -0,0 +1,296 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"max_vus": 8,
|
| 4 |
+
"duration_secs": 120,
|
| 5 |
+
"benchmark_kind": "Rate",
|
| 6 |
+
"warmup_duration_secs": 30,
|
| 7 |
+
"rates": [
|
| 8 |
+
1.0,
|
| 9 |
+
10.0,
|
| 10 |
+
30.0,
|
| 11 |
+
100.0
|
| 12 |
+
],
|
| 13 |
+
"num_rates": 10,
|
| 14 |
+
"prompt_options": {
|
| 15 |
+
"num_tokens": 200,
|
| 16 |
+
"min_tokens": 180,
|
| 17 |
+
"max_tokens": 220,
|
| 18 |
+
"variance": 10
|
| 19 |
+
},
|
| 20 |
+
"decode_options": {
|
| 21 |
+
"num_tokens": 200,
|
| 22 |
+
"min_tokens": 180,
|
| 23 |
+
"max_tokens": 220,
|
| 24 |
+
"variance": 10
|
| 25 |
+
},
|
| 26 |
+
"tokenizer": "microsoft/phi-4",
|
| 27 |
+
"model_name": "phi-4",
|
| 28 |
+
"profile": null,
|
| 29 |
+
"meta": null,
|
| 30 |
+
"run_id": "ik_llama.cpp -np 16, max-vus 8: unsloth/phi-4-GGUF:Q8_0 (200 tokens)"
|
| 31 |
+
},
|
| 32 |
+
"results": [
|
| 33 |
+
{
|
| 34 |
+
"id": "warmup",
|
| 35 |
+
"executor_type": "ConstantVUs",
|
| 36 |
+
"config": {
|
| 37 |
+
"max_vus": 1,
|
| 38 |
+
"duration_secs": 30,
|
| 39 |
+
"rate": null
|
| 40 |
+
},
|
| 41 |
+
"total_requests": 11,
|
| 42 |
+
"total_tokens": 2042,
|
| 43 |
+
"token_throughput_secs": 69.33786408424527,
|
| 44 |
+
"duration_ms": 29449,
|
| 45 |
+
"time_to_first_token_ms": {
|
| 46 |
+
"p50": 60.733,
|
| 47 |
+
"p60": 62.158,
|
| 48 |
+
"p70": 65.977,
|
| 49 |
+
"p80": 69.027,
|
| 50 |
+
"p90": 72.715,
|
| 51 |
+
"p95": 73.763,
|
| 52 |
+
"p99": 74.601,
|
| 53 |
+
"avg": 61.471
|
| 54 |
+
},
|
| 55 |
+
"inter_token_latency_ms": {
|
| 56 |
+
"p50": 14.12,
|
| 57 |
+
"p60": 14.369,
|
| 58 |
+
"p70": 14.541,
|
| 59 |
+
"p80": 14.79,
|
| 60 |
+
"p90": 15.038,
|
| 61 |
+
"p95": 15.163,
|
| 62 |
+
"p99": 15.263,
|
| 63 |
+
"avg": 13.957
|
| 64 |
+
},
|
| 65 |
+
"failed_requests": 0,
|
| 66 |
+
"successful_requests": 11,
|
| 67 |
+
"request_rate": 0.3735144490336425,
|
| 68 |
+
"total_tokens_sent": 2200,
|
| 69 |
+
"e2e_latency_ms": {
|
| 70 |
+
"p50": 2975.459,
|
| 71 |
+
"p60": 3009.442,
|
| 72 |
+
"p70": 3024.936,
|
| 73 |
+
"p80": 3054.238,
|
| 74 |
+
"p90": 3092.173,
|
| 75 |
+
"p95": 3125.964,
|
| 76 |
+
"p99": 3152.996,
|
| 77 |
+
"avg": 2677.072
|
| 78 |
+
}
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"id": "[email protected]/s",
|
| 82 |
+
"executor_type": "ConstantArrivalRate",
|
| 83 |
+
"config": {
|
| 84 |
+
"max_vus": 8,
|
| 85 |
+
"duration_secs": 120,
|
| 86 |
+
"rate": 1.0
|
| 87 |
+
},
|
| 88 |
+
"total_requests": 115,
|
| 89 |
+
"total_tokens": 21668,
|
| 90 |
+
"token_throughput_secs": 182.3320833907104,
|
| 91 |
+
"duration_ms": 118838,
|
| 92 |
+
"time_to_first_token_ms": {
|
| 93 |
+
"p50": 104.757,
|
| 94 |
+
"p60": 106.656,
|
| 95 |
+
"p70": 109.218,
|
| 96 |
+
"p80": 111.488,
|
| 97 |
+
"p90": 114.362,
|
| 98 |
+
"p95": 117.768,
|
| 99 |
+
"p99": 121.927,
|
| 100 |
+
"avg": 105.067
|
| 101 |
+
},
|
| 102 |
+
"inter_token_latency_ms": {
|
| 103 |
+
"p50": 22.886,
|
| 104 |
+
"p60": 23.016,
|
| 105 |
+
"p70": 23.167,
|
| 106 |
+
"p80": 23.27,
|
| 107 |
+
"p90": 23.402,
|
| 108 |
+
"p95": 23.502,
|
| 109 |
+
"p99": 23.78,
|
| 110 |
+
"avg": 22.916
|
| 111 |
+
},
|
| 112 |
+
"failed_requests": 0,
|
| 113 |
+
"successful_requests": 115,
|
| 114 |
+
"request_rate": 0.9677030455017397,
|
| 115 |
+
"total_tokens_sent": 23000,
|
| 116 |
+
"e2e_latency_ms": {
|
| 117 |
+
"p50": 4553.56,
|
| 118 |
+
"p60": 4622.268,
|
| 119 |
+
"p70": 4682.461,
|
| 120 |
+
"p80": 4762.847,
|
| 121 |
+
"p90": 4849.725,
|
| 122 |
+
"p95": 4981.403,
|
| 123 |
+
"p99": 5146.162,
|
| 124 |
+
"avg": 4403.649
|
| 125 |
+
}
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"id": "[email protected]/s",
|
| 129 |
+
"executor_type": "ConstantArrivalRate",
|
| 130 |
+
"config": {
|
| 131 |
+
"max_vus": 8,
|
| 132 |
+
"duration_secs": 120,
|
| 133 |
+
"rate": 10.0
|
| 134 |
+
},
|
| 135 |
+
"total_requests": 180,
|
| 136 |
+
"total_tokens": 34169,
|
| 137 |
+
"token_throughput_secs": 287.4269957193569,
|
| 138 |
+
"duration_ms": 118878,
|
| 139 |
+
"time_to_first_token_ms": {
|
| 140 |
+
"p50": 107.982,
|
| 141 |
+
"p60": 110.085,
|
| 142 |
+
"p70": 112.718,
|
| 143 |
+
"p80": 114.259,
|
| 144 |
+
"p90": 116.621,
|
| 145 |
+
"p95": 118.596,
|
| 146 |
+
"p99": 126.345,
|
| 147 |
+
"avg": 108.063
|
| 148 |
+
},
|
| 149 |
+
"inter_token_latency_ms": {
|
| 150 |
+
"p50": 26.586,
|
| 151 |
+
"p60": 26.723,
|
| 152 |
+
"p70": 26.832,
|
| 153 |
+
"p80": 26.954,
|
| 154 |
+
"p90": 27.378,
|
| 155 |
+
"p95": 27.765,
|
| 156 |
+
"p99": 28.187,
|
| 157 |
+
"avg": 26.571
|
| 158 |
+
},
|
| 159 |
+
"failed_requests": 0,
|
| 160 |
+
"successful_requests": 180,
|
| 161 |
+
"request_rate": 1.514146133322141,
|
| 162 |
+
"total_tokens_sent": 36000,
|
| 163 |
+
"e2e_latency_ms": {
|
| 164 |
+
"p50": 5340.395,
|
| 165 |
+
"p60": 5423.095,
|
| 166 |
+
"p70": 5515.212,
|
| 167 |
+
"p80": 5624.709,
|
| 168 |
+
"p90": 5733.59,
|
| 169 |
+
"p95": 5808.81,
|
| 170 |
+
"p99": 6051.086,
|
| 171 |
+
"avg": 5140.977
|
| 172 |
+
}
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"id": "[email protected]/s",
|
| 176 |
+
"executor_type": "ConstantArrivalRate",
|
| 177 |
+
"config": {
|
| 178 |
+
"max_vus": 8,
|
| 179 |
+
"duration_secs": 120,
|
| 180 |
+
"rate": 30.0
|
| 181 |
+
},
|
| 182 |
+
"total_requests": 177,
|
| 183 |
+
"total_tokens": 33800,
|
| 184 |
+
"token_throughput_secs": 285.37390254245355,
|
| 185 |
+
"duration_ms": 118441,
|
| 186 |
+
"time_to_first_token_ms": {
|
| 187 |
+
"p50": 114.476,
|
| 188 |
+
"p60": 116.168,
|
| 189 |
+
"p70": 119.348,
|
| 190 |
+
"p80": 122.294,
|
| 191 |
+
"p90": 129.873,
|
| 192 |
+
"p95": 182.17,
|
| 193 |
+
"p99": 550.519,
|
| 194 |
+
"avg": 128.85
|
| 195 |
+
},
|
| 196 |
+
"inter_token_latency_ms": {
|
| 197 |
+
"p50": 26.808,
|
| 198 |
+
"p60": 26.923,
|
| 199 |
+
"p70": 27.109,
|
| 200 |
+
"p80": 27.26,
|
| 201 |
+
"p90": 27.46,
|
| 202 |
+
"p95": 27.966,
|
| 203 |
+
"p99": 29.407,
|
| 204 |
+
"avg": 26.861
|
| 205 |
+
},
|
| 206 |
+
"failed_requests": 0,
|
| 207 |
+
"successful_requests": 177,
|
| 208 |
+
"request_rate": 1.4944136316572272,
|
| 209 |
+
"total_tokens_sent": 35400,
|
| 210 |
+
"e2e_latency_ms": {
|
| 211 |
+
"p50": 5407.958,
|
| 212 |
+
"p60": 5455.363,
|
| 213 |
+
"p70": 5538.56,
|
| 214 |
+
"p80": 5674.894,
|
| 215 |
+
"p90": 5823.38,
|
| 216 |
+
"p95": 5868.327,
|
| 217 |
+
"p99": 6157.756,
|
| 218 |
+
"avg": 5240.665
|
| 219 |
+
}
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"id": "[email protected]/s",
|
| 223 |
+
"executor_type": "ConstantArrivalRate",
|
| 224 |
+
"config": {
|
| 225 |
+
"max_vus": 8,
|
| 226 |
+
"duration_secs": 120,
|
| 227 |
+
"rate": 100.0
|
| 228 |
+
},
|
| 229 |
+
"total_requests": 182,
|
| 230 |
+
"total_tokens": 33812,
|
| 231 |
+
"token_throughput_secs": 285.72660230737114,
|
| 232 |
+
"duration_ms": 118336,
|
| 233 |
+
"time_to_first_token_ms": {
|
| 234 |
+
"p50": 116.911,
|
| 235 |
+
"p60": 118.312,
|
| 236 |
+
"p70": 119.797,
|
| 237 |
+
"p80": 122.559,
|
| 238 |
+
"p90": 179.079,
|
| 239 |
+
"p95": 196.698,
|
| 240 |
+
"p99": 649.72,
|
| 241 |
+
"avg": 141.58
|
| 242 |
+
},
|
| 243 |
+
"inter_token_latency_ms": {
|
| 244 |
+
"p50": 26.804,
|
| 245 |
+
"p60": 26.937,
|
| 246 |
+
"p70": 27.148,
|
| 247 |
+
"p80": 27.363,
|
| 248 |
+
"p90": 27.66,
|
| 249 |
+
"p95": 27.979,
|
| 250 |
+
"p99": 30.34,
|
| 251 |
+
"avg": 26.903
|
| 252 |
+
},
|
| 253 |
+
"failed_requests": 0,
|
| 254 |
+
"successful_requests": 182,
|
| 255 |
+
"request_rate": 1.5379818295262495,
|
| 256 |
+
"total_tokens_sent": 36400,
|
| 257 |
+
"e2e_latency_ms": {
|
| 258 |
+
"p50": 5386.275,
|
| 259 |
+
"p60": 5437.527,
|
| 260 |
+
"p70": 5544.241,
|
| 261 |
+
"p80": 5702.711,
|
| 262 |
+
"p90": 5834.793,
|
| 263 |
+
"p95": 5923.803,
|
| 264 |
+
"p99": 6090.788,
|
| 265 |
+
"avg": 5111.129
|
| 266 |
+
}
|
| 267 |
+
}
|
| 268 |
+
],
|
| 269 |
+
"start_time": "2025-05-28T14:01:37.984694850+00:00",
|
| 270 |
+
"end_time": "2025-05-28T14:09:35.158748747+00:00",
|
| 271 |
+
"system": {
|
| 272 |
+
"cpu": [
|
| 273 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu0@4699MHz",
|
| 274 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu1@4699MHz",
|
| 275 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu2@4699MHz",
|
| 276 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu3@4699MHz",
|
| 277 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu4@4699MHz",
|
| 278 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu5@4699MHz",
|
| 279 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu6@4699MHz",
|
| 280 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu7@4699MHz",
|
| 281 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu8@4699MHz",
|
| 282 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu9@4699MHz",
|
| 283 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu10@4699MHz",
|
| 284 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu11@4699MHz",
|
| 285 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu12@4699MHz",
|
| 286 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu13@4699MHz",
|
| 287 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu14@4699MHz",
|
| 288 |
+
"AMD Ryzen 7 9800X3D 8-Core Processor cpu15@4699MHz"
|
| 289 |
+
],
|
| 290 |
+
"memory": "83.47 GB",
|
| 291 |
+
"os_name": "Debian GNU/Linux",
|
| 292 |
+
"os_version": "11",
|
| 293 |
+
"kernel": "5.15.167.4-microsoft-standard-WSL2",
|
| 294 |
+
"hostname": "computer"
|
| 295 |
+
}
|
| 296 |
+
}
|