Upload microsoft_phi-4_2025-05-27-13-36-57.json
Browse filessudo docker run --runtime=nvidia --gpus all --net=host --ipc=host \
-v ~/.cache/huggingface:/root/.cache/huggingface \
-e HUGGING_FACE_HUB_TOKEN=$HUGGING_FACE_HUB_TOKEN \
-e VLLM_ATTENTION_BACKEND=FLASH_INFER \
myvllm-kaln27-updated \
--model RedHatAI/phi-4-FP8-dynamic \
--max-model-len 16384 \
--enable-prefix-caching \
--enable-chunked-prefill \
--disable-log-requests
results/microsoft_phi-4_2025-05-27-13-36-57.json
ADDED
@@ -0,0 +1,296 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"config": {
|
3 |
+
"max_vus": 800,
|
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": "vLLM FLASHINFER --enable-prefix-caching --enable-chunked-prefill: RedHatAI/phi-4-FP8-dynamic (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": 7,
|
42 |
+
"total_tokens": 1378,
|
43 |
+
"token_throughput_secs": 40.36809975645149,
|
44 |
+
"duration_ms": 34135,
|
45 |
+
"time_to_first_token_ms": {
|
46 |
+
"p50": 54.285,
|
47 |
+
"p60": 55.398,
|
48 |
+
"p70": 56.232,
|
49 |
+
"p80": 56.504,
|
50 |
+
"p90": 453.069,
|
51 |
+
"p95": 750.424,
|
52 |
+
"p99": 988.308,
|
53 |
+
"avg": 196.458
|
54 |
+
},
|
55 |
+
"inter_token_latency_ms": {
|
56 |
+
"p50": 23.781,
|
57 |
+
"p60": 23.784,
|
58 |
+
"p70": 23.828,
|
59 |
+
"p80": 23.955,
|
60 |
+
"p90": 24.363,
|
61 |
+
"p95": 24.638,
|
62 |
+
"p99": 24.857,
|
63 |
+
"avg": 23.895
|
64 |
+
},
|
65 |
+
"failed_requests": 0,
|
66 |
+
"successful_requests": 7,
|
67 |
+
"request_rate": 0.20506291603422383,
|
68 |
+
"total_tokens_sent": 1400,
|
69 |
+
"e2e_latency_ms": {
|
70 |
+
"p50": 4775.092,
|
71 |
+
"p60": 4786.27,
|
72 |
+
"p70": 4827.876,
|
73 |
+
"p80": 4930.341,
|
74 |
+
"p90": 5248.136,
|
75 |
+
"p95": 5460.866,
|
76 |
+
"p99": 5631.05,
|
77 |
+
"avg": 4876.39
|
78 |
+
}
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"id": "[email protected]/s",
|
82 |
+
"executor_type": "ConstantArrivalRate",
|
83 |
+
"config": {
|
84 |
+
"max_vus": 800,
|
85 |
+
"duration_secs": 120,
|
86 |
+
"rate": 1.0
|
87 |
+
},
|
88 |
+
"total_requests": 115,
|
89 |
+
"total_tokens": 22432,
|
90 |
+
"token_throughput_secs": 188.6951002433708,
|
91 |
+
"duration_ms": 118879,
|
92 |
+
"time_to_first_token_ms": {
|
93 |
+
"p50": 46.012,
|
94 |
+
"p60": 49.416,
|
95 |
+
"p70": 51.301,
|
96 |
+
"p80": 54.8,
|
97 |
+
"p90": 56.788,
|
98 |
+
"p95": 57.91,
|
99 |
+
"p99": 63.157,
|
100 |
+
"avg": 46.748
|
101 |
+
},
|
102 |
+
"inter_token_latency_ms": {
|
103 |
+
"p50": 24.164,
|
104 |
+
"p60": 24.225,
|
105 |
+
"p70": 24.262,
|
106 |
+
"p80": 24.34,
|
107 |
+
"p90": 24.595,
|
108 |
+
"p95": 24.658,
|
109 |
+
"p99": 24.724,
|
110 |
+
"avg": 24.218
|
111 |
+
},
|
112 |
+
"failed_requests": 0,
|
113 |
+
"successful_requests": 115,
|
114 |
+
"request_rate": 0.967365216119278,
|
115 |
+
"total_tokens_sent": 23000,
|
116 |
+
"e2e_latency_ms": {
|
117 |
+
"p50": 4845.831,
|
118 |
+
"p60": 4937.108,
|
119 |
+
"p70": 4994.642,
|
120 |
+
"p80": 5083.534,
|
121 |
+
"p90": 5155.065,
|
122 |
+
"p95": 5222.709,
|
123 |
+
"p99": 5315.988,
|
124 |
+
"avg": 4748.625
|
125 |
+
}
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"id": "[email protected]/s",
|
129 |
+
"executor_type": "ConstantArrivalRate",
|
130 |
+
"config": {
|
131 |
+
"max_vus": 800,
|
132 |
+
"duration_secs": 120,
|
133 |
+
"rate": 10.0
|
134 |
+
},
|
135 |
+
"total_requests": 1142,
|
136 |
+
"total_tokens": 214058,
|
137 |
+
"token_throughput_secs": 1785.635788036227,
|
138 |
+
"duration_ms": 119877,
|
139 |
+
"time_to_first_token_ms": {
|
140 |
+
"p50": 57.214,
|
141 |
+
"p60": 60.195,
|
142 |
+
"p70": 62.457,
|
143 |
+
"p80": 65.608,
|
144 |
+
"p90": 68.347,
|
145 |
+
"p95": 70.005,
|
146 |
+
"p99": 73.831,
|
147 |
+
"avg": 57.823
|
148 |
+
},
|
149 |
+
"inter_token_latency_ms": {
|
150 |
+
"p50": 29.792,
|
151 |
+
"p60": 29.941,
|
152 |
+
"p70": 30.098,
|
153 |
+
"p80": 30.274,
|
154 |
+
"p90": 30.81,
|
155 |
+
"p95": 30.992,
|
156 |
+
"p99": 31.408,
|
157 |
+
"avg": 29.837
|
158 |
+
},
|
159 |
+
"failed_requests": 0,
|
160 |
+
"successful_requests": 1142,
|
161 |
+
"request_rate": 9.526371684017281,
|
162 |
+
"total_tokens_sent": 228400,
|
163 |
+
"e2e_latency_ms": {
|
164 |
+
"p50": 5924.829,
|
165 |
+
"p60": 6019.926,
|
166 |
+
"p70": 6100.029,
|
167 |
+
"p80": 6204.284,
|
168 |
+
"p90": 6362.309,
|
169 |
+
"p95": 6475.238,
|
170 |
+
"p99": 6682.944,
|
171 |
+
"avg": 5622.484
|
172 |
+
}
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"id": "[email protected]/s",
|
176 |
+
"executor_type": "ConstantArrivalRate",
|
177 |
+
"config": {
|
178 |
+
"max_vus": 800,
|
179 |
+
"duration_secs": 120,
|
180 |
+
"rate": 30.0
|
181 |
+
},
|
182 |
+
"total_requests": 1464,
|
183 |
+
"total_tokens": 268954,
|
184 |
+
"token_throughput_secs": 2261.2375594297523,
|
185 |
+
"duration_ms": 118941,
|
186 |
+
"time_to_first_token_ms": {
|
187 |
+
"p50": 21462.473,
|
188 |
+
"p60": 29737.974,
|
189 |
+
"p70": 36838.741,
|
190 |
+
"p80": 37849.964,
|
191 |
+
"p90": 39658.621,
|
192 |
+
"p95": 45477.778,
|
193 |
+
"p99": 46753.911,
|
194 |
+
"avg": 22718.219
|
195 |
+
},
|
196 |
+
"inter_token_latency_ms": {
|
197 |
+
"p50": 87.204,
|
198 |
+
"p60": 90.059,
|
199 |
+
"p70": 93.033,
|
200 |
+
"p80": 97.634,
|
201 |
+
"p90": 118.914,
|
202 |
+
"p95": 126.932,
|
203 |
+
"p99": 188.85,
|
204 |
+
"avg": 92.276
|
205 |
+
},
|
206 |
+
"failed_requests": 0,
|
207 |
+
"successful_requests": 1464,
|
208 |
+
"request_rate": 12.308617038620573,
|
209 |
+
"total_tokens_sent": 292800,
|
210 |
+
"e2e_latency_ms": {
|
211 |
+
"p50": 39522.38,
|
212 |
+
"p60": 46133.615,
|
213 |
+
"p70": 51035.266,
|
214 |
+
"p80": 54406.736,
|
215 |
+
"p90": 58251.04,
|
216 |
+
"p95": 60717.414,
|
217 |
+
"p99": 63893.309,
|
218 |
+
"avg": 39205.997
|
219 |
+
}
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"id": "[email protected]/s",
|
223 |
+
"executor_type": "ConstantArrivalRate",
|
224 |
+
"config": {
|
225 |
+
"max_vus": 800,
|
226 |
+
"duration_secs": 120,
|
227 |
+
"rate": 100.0
|
228 |
+
},
|
229 |
+
"total_requests": 1353,
|
230 |
+
"total_tokens": 244634,
|
231 |
+
"token_throughput_secs": 2040.3733554203798,
|
232 |
+
"duration_ms": 119896,
|
233 |
+
"time_to_first_token_ms": {
|
234 |
+
"p50": 38485.207,
|
235 |
+
"p60": 42311.807,
|
236 |
+
"p70": 43242.686,
|
237 |
+
"p80": 45331.449,
|
238 |
+
"p90": 52807.775,
|
239 |
+
"p95": 53519.524,
|
240 |
+
"p99": 54382.196,
|
241 |
+
"avg": 32240.618
|
242 |
+
},
|
243 |
+
"inter_token_latency_ms": {
|
244 |
+
"p50": 96.979,
|
245 |
+
"p60": 99.558,
|
246 |
+
"p70": 102.098,
|
247 |
+
"p80": 107.625,
|
248 |
+
"p90": 135.279,
|
249 |
+
"p95": 155.304,
|
250 |
+
"p99": 243.309,
|
251 |
+
"avg": 106.949
|
252 |
+
},
|
253 |
+
"failed_requests": 0,
|
254 |
+
"successful_requests": 1353,
|
255 |
+
"request_rate": 11.2847157381385,
|
256 |
+
"total_tokens_sent": 270600,
|
257 |
+
"e2e_latency_ms": {
|
258 |
+
"p50": 55925.819,
|
259 |
+
"p60": 59369.223,
|
260 |
+
"p70": 62535.903,
|
261 |
+
"p80": 66361.009,
|
262 |
+
"p90": 70934.25,
|
263 |
+
"p95": 72596.577,
|
264 |
+
"p99": 75860.246,
|
265 |
+
"avg": 50326.788
|
266 |
+
}
|
267 |
+
}
|
268 |
+
],
|
269 |
+
"start_time": "2025-05-27T13:26:47.155111664+00:00",
|
270 |
+
"end_time": "2025-05-27T13:36:57.064904817+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 |
+
}
|