Hugging Face's logo

Datasets:
optimum-benchmark
/
cpu

Dataset card Viewer Files Community
IlyasMoutawwakil HF staff commited on
Commit
92adc62
•
1 Parent(s): f6805bd

Upload cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json +91 -97
cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json CHANGED
@@ -3,7 +3,7 @@
3
  "name": "cpu_inference_transformers_text-generation_openai-community/gpt2",
4
  "backend": {
5
  "name": "pytorch",
6
- "version": "2.3.1+cpu",
7
  "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
  "task": "text-generation",
9
  "library": "transformers",
@@ -76,14 +76,14 @@
76
  "cpu_ram_mb": 16757.342208,
77
  "system": "Linux",
78
  "machine": "x86_64",
79
- "platform": "Linux-6.5.0-1023-azure-x86_64-with-glibc2.35",
80
  "processor": "x86_64",
81
  "python_version": "3.10.14",
82
  "optimum_benchmark_version": "0.3.1",
83
- "optimum_benchmark_commit": "328b924ef0be0164f0dc86652abfd3746f634c6b",
84
  "transformers_version": "4.42.4",
85
  "transformers_commit": null,
86
- "accelerate_version": "0.32.1",
87
  "accelerate_commit": null,
88
  "diffusers_version": "0.29.2",
89
  "diffusers_commit": null,
@@ -99,7 +99,7 @@
99
  "load": {
100
  "memory": {
101
  "unit": "MB",
102
- "max_ram": 1116.135424,
103
  "max_global_vram": null,
104
  "max_process_vram": null,
105
  "max_reserved": null,
@@ -108,31 +108,31 @@
108
  "latency": {
109
  "unit": "s",
110
  "count": 1,
111
- "total": 4.770297263000003,
112
- "mean": 4.770297263000003,
113
  "stdev": 0.0,
114
- "p50": 4.770297263000003,
115
- "p90": 4.770297263000003,
116
- "p95": 4.770297263000003,
117
- "p99": 4.770297263000003,
118
  "values": [
119
- 4.770297263000003
120
  ]
121
  },
122
  "throughput": null,
123
  "energy": {
124
  "unit": "kWh",
125
- "cpu": 6.811593373616537e-05,
126
- "ram": 2.8470770394358167e-06,
127
  "gpu": 0,
128
- "total": 7.096301077560119e-05
129
  },
130
  "efficiency": null
131
  },
132
  "prefill": {
133
  "memory": {
134
  "unit": "MB",
135
- "max_ram": 969.469952,
136
  "max_global_vram": null,
137
  "max_process_vram": null,
138
  "max_reserved": null,
@@ -140,52 +140,50 @@
140
  },
141
  "latency": {
142
  "unit": "s",
143
- "count": 15,
144
- "total": 0.6367758400002685,
145
- "mean": 0.04245172266668457,
146
- "stdev": 0.00396891351029044,
147
- "p50": 0.04103058800001236,
148
- "p90": 0.043928176200040525,
149
- "p95": 0.047996650900034854,
150
- "p99": 0.05499496698001052,
151
  "values": [
152
- 0.04103058800001236,
153
- 0.04240679200000841,
154
- 0.04344911100002946,
155
- 0.04150333100000125,
156
- 0.041329226000016206,
157
- 0.04083185600001116,
158
- 0.040988208000044324,
159
- 0.04088525600002413,
160
- 0.04112043600002835,
161
- 0.04085750399997323,
162
- 0.04025056900002255,
163
- 0.0442475530000479,
164
- 0.04079309300004752,
165
- 0.04033777099999725,
166
- 0.05674454600000445
167
  ]
168
  },
169
  "throughput": {
170
  "unit": "tokens/s",
171
- "value": 47.11234019178138
172
  },
173
  "energy": {
174
  "unit": "kWh",
175
- "cpu": 1.5672866503397626e-06,
176
- "ram": 6.55000627469235e-08,
177
  "gpu": 0.0,
178
- "total": 1.632786713086686e-06
179
  },
180
  "efficiency": {
181
  "unit": "tokens/kWh",
182
- "value": 1224899.727545626
183
  }
184
  },
185
  "decode": {
186
  "memory": {
187
  "unit": "MB",
188
- "max_ram": 970.387456,
189
  "max_global_vram": null,
190
  "max_process_vram": null,
191
  "max_reserved": null,
@@ -193,81 +191,77 @@
193
  },
194
  "latency": {
195
  "unit": "s",
196
- "count": 15,
197
- "total": 0.3832927019998351,
198
- "mean": 0.025552846799989006,
199
- "stdev": 0.0022979591211475248,
200
- "p50": 0.02466371700000991,
201
- "p90": 0.027832154800000806,
202
- "p95": 0.029826911500020963,
203
- "p99": 0.0323912775000133,
204
  "values": [
205
- 0.028453144000025077,
206
- 0.025418919999992795,
207
- 0.024850947000004453,
208
- 0.026900670999964404,
209
- 0.02477871299998924,
210
- 0.02438311199995269,
211
- 0.024762932999976783,
212
- 0.024516861999984485,
213
- 0.0243242720000012,
214
- 0.02466371700000991,
215
- 0.024517272999958095,
216
- 0.0243293030000018,
217
- 0.023845055999970555,
218
- 0.024515409999992244,
219
- 0.033032369000011386
220
  ]
221
  },
222
  "throughput": {
223
  "unit": "tokens/s",
224
- "value": 39.134582844226586
225
  },
226
  "energy": {
227
  "unit": "kWh",
228
- "cpu": 9.199737012386319e-07,
229
- "ram": 3.844739625948249e-08,
230
  "gpu": 0.0,
231
- "total": 9.584210974981153e-07
232
  },
233
  "efficiency": {
234
  "unit": "tokens/kWh",
235
- "value": 1043382.7078832292
236
  }
237
  },
238
  "per_token": {
239
  "memory": null,
240
  "latency": {
241
  "unit": "s",
242
- "count": 15,
243
- "total": 0.37797582099977944,
244
- "mean": 0.025198388066651964,
245
- "stdev": 0.002272434881375836,
246
- "p50": 0.02432994299999791,
247
- "p90": 0.027298450799980854,
248
- "p95": 0.029359821199989252,
249
- "p99": 0.032034663439992526,
250
  "values": [
251
- 0.027926869999987503,
252
- 0.025077090999957363,
253
- 0.024529315999984647,
254
- 0.026355821999970885,
255
- 0.024452732999975524,
256
- 0.02405812399996421,
257
- 0.024437605000002804,
258
- 0.02419282599998951,
259
- 0.023969577999991998,
260
- 0.02432994299999791,
261
- 0.02418975099999443,
262
- 0.024029752999979337,
263
- 0.02352975699994886,
264
- 0.02419327800004112,
265
- 0.032703373999993346
266
  ]
267
  },
268
  "throughput": {
269
  "unit": "tokens/s",
270
- "value": 39.68507816273452
271
  },
272
  "energy": null,
273
  "efficiency": null
 
3
  "name": "cpu_inference_transformers_text-generation_openai-community/gpt2",
4
  "backend": {
5
  "name": "pytorch",
6
+ "version": "2.4.0+cpu",
7
  "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
  "task": "text-generation",
9
  "library": "transformers",
 
76
  "cpu_ram_mb": 16757.342208,
77
  "system": "Linux",
78
  "machine": "x86_64",
79
+ "platform": "Linux-6.5.0-1024-azure-x86_64-with-glibc2.35",
80
  "processor": "x86_64",
81
  "python_version": "3.10.14",
82
  "optimum_benchmark_version": "0.3.1",
83
+ "optimum_benchmark_commit": "b0801269b3611e452bb077a62163b08a99ceb2a9",
84
  "transformers_version": "4.42.4",
85
  "transformers_commit": null,
86
+ "accelerate_version": "0.33.0",
87
  "accelerate_commit": null,
88
  "diffusers_version": "0.29.2",
89
  "diffusers_commit": null,
 
99
  "load": {
100
  "memory": {
101
  "unit": "MB",
102
+ "max_ram": 1120.772096,
103
  "max_global_vram": null,
104
  "max_process_vram": null,
105
  "max_reserved": null,
 
108
  "latency": {
109
  "unit": "s",
110
  "count": 1,
111
+ "total": 4.834404436999989,
112
+ "mean": 4.834404436999989,
113
  "stdev": 0.0,
114
+ "p50": 4.834404436999989,
115
+ "p90": 4.834404436999989,
116
+ "p95": 4.834404436999989,
117
+ "p99": 4.834404436999989,
118
  "values": [
119
+ 4.834404436999989
120
  ]
121
  },
122
  "throughput": null,
123
  "energy": {
124
  "unit": "kWh",
125
+ "cpu": 6.756949822107951e-05,
126
+ "ram": 2.824225813326393e-06,
127
  "gpu": 0,
128
+ "total": 7.03937240344059e-05
129
  },
130
  "efficiency": null
131
  },
132
  "prefill": {
133
  "memory": {
134
  "unit": "MB",
135
+ "max_ram": 974.503936,
136
  "max_global_vram": null,
137
  "max_process_vram": null,
138
  "max_reserved": null,
 
140
  },
141
  "latency": {
142
  "unit": "s",
143
+ "count": 13,
144
+ "total": 0.624935255000139,
145
+ "mean": 0.04807194269231839,
146
+ "stdev": 0.0008469223243709784,
147
+ "p50": 0.04825908699996262,
148
+ "p90": 0.04896189200001118,
149
+ "p95": 0.04909340240002393,
150
+ "p99": 0.0491845716800367,
151
  "values": [
152
+ 0.04638163900000336,
153
+ 0.04825908699996262,
154
+ 0.0490174280000133,
155
+ 0.047192688999984966,
156
+ 0.04920736400003989,
157
+ 0.04796389400001999,
158
+ 0.04654725000000326,
159
+ 0.04873974800000269,
160
+ 0.04834172200003195,
161
+ 0.04849652199999355,
162
+ 0.047984503000009227,
163
+ 0.04873014000003195,
164
+ 0.048073269000042274
 
 
165
  ]
166
  },
167
  "throughput": {
168
  "unit": "tokens/s",
169
+ "value": 41.604309873659176
170
  },
171
  "energy": {
172
  "unit": "kWh",
173
+ "cpu": 1.723505678960091e-06,
174
+ "ram": 7.202735670489345e-08,
175
  "gpu": 0.0,
176
+ "total": 1.7955330356649845e-06
177
  },
178
  "efficiency": {
179
  "unit": "tokens/kWh",
180
+ "value": 1113875.356383677
181
  }
182
  },
183
  "decode": {
184
  "memory": {
185
  "unit": "MB",
186
+ "max_ram": 975.290368,
187
  "max_global_vram": null,
188
  "max_process_vram": null,
189
  "max_reserved": null,
 
191
  },
192
  "latency": {
193
  "unit": "s",
194
+ "count": 13,
195
+ "total": 0.38288725999984763,
196
+ "mean": 0.029452866153834432,
197
+ "stdev": 0.0012264574497014423,
198
+ "p50": 0.030012737999982164,
199
+ "p90": 0.03021878959998503,
200
+ "p95": 0.03050145339997243,
201
+ "p99": 0.03082699947996616,
202
  "values": [
203
+ 0.03017328800001451,
204
+ 0.030088269000032142,
205
+ 0.02921427199999016,
206
+ 0.029231182999978955,
207
+ 0.030908385999964594,
208
+ 0.029763671000011982,
209
+ 0.028035321999993812,
210
+ 0.03009783699997115,
211
+ 0.03023016499997766,
212
+ 0.029081232000010004,
213
+ 0.030117623999956322,
214
+ 0.030012737999982164,
215
+ 0.025933272999964174
 
 
216
  ]
217
  },
218
  "throughput": {
219
  "unit": "tokens/s",
220
+ "value": 33.95255303089785
221
  },
222
  "energy": {
223
  "unit": "kWh",
224
+ "cpu": 1.012130290413824e-06,
225
+ "ram": 4.230164102523302e-08,
226
  "gpu": 0.0,
227
+ "total": 1.0544319314390573e-06
228
  },
229
  "efficiency": {
230
  "unit": "tokens/kWh",
231
+ "value": 948377.9561144642
232
  }
233
  },
234
  "per_token": {
235
  "memory": null,
236
  "latency": {
237
  "unit": "s",
238
+ "count": 13,
239
+ "total": 0.3764351199998828,
240
+ "mean": 0.028956547692298676,
241
+ "stdev": 0.0011848917693045647,
242
+ "p50": 0.029534071000000495,
243
+ "p90": 0.029664990599974315,
244
+ "p95": 0.029957622599977182,
245
+ "p99": 0.030304674119988703,
246
  "values": [
247
+ 0.02965130100000124,
248
+ 0.029608870000004117,
249
+ 0.02871185100002549,
250
+ 0.028710967999984405,
251
+ 0.030391436999991583,
252
+ 0.029231543999969745,
253
+ 0.027562686000010217,
254
+ 0.029564206999964426,
255
+ 0.029668412999967586,
256
+ 0.02856910200000584,
257
+ 0.029638755999997102,
258
+ 0.029534071000000495,
259
+ 0.025591913999960525
 
 
260
  ]
261
  },
262
  "throughput": {
263
  "unit": "tokens/s",
264
+ "value": 34.534503581929464
265
  },
266
  "energy": null,
267
  "efficiency": null