Hugging Face's logo

Datasets:
optimum-benchmark
/
cpu

Dataset card Viewer Files Community
IlyasMoutawwakil HF staff commited on
Commit
24766a7
•
1 Parent(s): 0da2aed

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 +82 -81
cpu_inference_transformers_text-generation_openai-community/gpt2/benchmark.json CHANGED
@@ -7,6 +7,7 @@
7
  "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
  "task": "text-generation",
9
  "library": "transformers",
 
10
  "model": "openai-community/gpt2",
11
  "processor": "openai-community/gpt2",
12
  "device": "cpu",
@@ -73,14 +74,14 @@
73
  "environment": {
74
  "cpu": " AMD EPYC 7763 64-Core Processor",
75
  "cpu_count": 4,
76
- "cpu_ram_mb": 16757.338112,
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": "c31e63ddb94c29460ba90ff599147e94ff80543a",
84
  "transformers_version": "4.42.4",
85
  "transformers_commit": null,
86
  "accelerate_version": "0.32.1",
@@ -99,7 +100,7 @@
99
  "prefill": {
100
  "memory": {
101
  "unit": "MB",
102
- "max_ram": 961.76128,
103
  "max_global_vram": null,
104
  "max_process_vram": null,
105
  "max_reserved": null,
@@ -108,51 +109,51 @@
108
  "latency": {
109
  "unit": "s",
110
  "count": 15,
111
- "total": 0.658391876999957,
112
- "mean": 0.04389279179999714,
113
- "stdev": 0.003143103223363988,
114
- "p50": 0.046014401000007865,
115
- "p90": 0.0468165909999982,
116
- "p95": 0.047008504299998324,
117
- "p99": 0.04702359685999568,
118
  "values": [
119
- 0.046014401000007865,
120
- 0.04654084899999589,
121
- 0.04700041899999974,
122
- 0.04616212700000233,
123
- 0.046489452000002984,
124
- 0.04702736999999502,
125
- 0.046293854999987616,
126
- 0.046441020000003164,
127
- 0.045928158999998914,
128
- 0.0400730220000014,
129
- 0.03958767199998192,
130
- 0.041238015999994104,
131
- 0.03968561499999623,
132
- 0.04002645399998528,
133
- 0.039883446000004596
134
  ]
135
  },
136
  "throughput": {
137
  "unit": "tokens/s",
138
- "value": 45.56556823984321
139
  },
140
  "energy": {
141
  "unit": "kWh",
142
- "cpu": 1.5671390427483453e-06,
143
- "ram": 6.549279102751674e-08,
144
  "gpu": 0.0,
145
- "total": 1.632631833775862e-06
146
  },
147
  "efficiency": {
148
  "unit": "tokens/kWh",
149
- "value": 1225015.927427134
150
  }
151
  },
152
  "decode": {
153
  "memory": {
154
  "unit": "MB",
155
- "max_ram": 961.76128,
156
  "max_global_vram": null,
157
  "max_process_vram": null,
158
  "max_reserved": null,
@@ -161,45 +162,45 @@
161
  "latency": {
162
  "unit": "s",
163
  "count": 15,
164
- "total": 0.4034822600000325,
165
- "mean": 0.0268988173333355,
166
- "stdev": 0.002157618476705122,
167
- "p50": 0.02838440700000433,
168
- "p90": 0.029005317200000037,
169
- "p95": 0.029413272400000776,
170
- "p99": 0.029847836880002206,
171
  "values": [
172
- 0.02838440700000433,
173
- 0.029956478000002562,
174
- 0.028708824999995386,
175
- 0.028650846999994428,
176
- 0.028693436999986943,
177
- 0.02874258800000007,
178
- 0.029180470000000014,
179
- 0.02866504300001793,
180
- 0.025008507999984886,
181
- 0.024309396999996125,
182
- 0.025252976000018634,
183
- 0.02449262000001795,
184
- 0.02511789299998668,
185
- 0.02404868800002191,
186
- 0.02427008300000466
187
  ]
188
  },
189
  "throughput": {
190
  "unit": "tokens/s",
191
- "value": 37.17635565935115
192
  },
193
  "energy": {
194
  "unit": "kWh",
195
- "cpu": 9.477513730525971e-07,
196
- "ram": 3.961036716685909e-08,
197
  "gpu": 0.0,
198
- "total": 9.873617402194565e-07
199
  },
200
  "efficiency": {
201
  "unit": "tokens/kWh",
202
- "value": 1012800.0298834088
203
  }
204
  },
205
  "per_token": {
@@ -207,34 +208,34 @@
207
  "latency": {
208
  "unit": "s",
209
  "count": 15,
210
- "total": 0.3974515710000901,
211
- "mean": 0.026496771400006006,
212
- "stdev": 0.002072399511716262,
213
- "p50": 0.0279198860000065,
214
- "p90": 0.02853342440001825,
215
- "p95": 0.028929358400014846,
216
- "p99": 0.02934266528000819,
217
  "values": [
218
- 0.0279198860000065,
219
- 0.029445992000006527,
220
- 0.028212404000015567,
221
- 0.028183249999983673,
222
- 0.028212765000006357,
223
- 0.028271645000017998,
224
- 0.028707944000018415,
225
- 0.028176397000009956,
226
- 0.02469757500000469,
227
- 0.024024902999997266,
228
- 0.024929678999995986,
229
- 0.024184462000022222,
230
- 0.024779147999993256,
231
- 0.023740991000011036,
232
- 0.02396453000000065
233
  ]
234
  },
235
  "throughput": {
236
  "unit": "tokens/s",
237
- "value": 37.74044712480605
238
  },
239
  "energy": null,
240
  "efficiency": null
 
7
  "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
  "task": "text-generation",
9
  "library": "transformers",
10
+ "model_type": "gpt2",
11
  "model": "openai-community/gpt2",
12
  "processor": "openai-community/gpt2",
13
  "device": "cpu",
 
74
  "environment": {
75
  "cpu": " AMD EPYC 7763 64-Core Processor",
76
  "cpu_count": 4,
77
+ "cpu_ram_mb": 16757.342208,
78
  "system": "Linux",
79
  "machine": "x86_64",
80
  "platform": "Linux-6.5.0-1023-azure-x86_64-with-glibc2.35",
81
  "processor": "x86_64",
82
  "python_version": "3.10.14",
83
  "optimum_benchmark_version": "0.3.1",
84
+ "optimum_benchmark_commit": "402644384abc46743e46dac65be6512bc68c9d92",
85
  "transformers_version": "4.42.4",
86
  "transformers_commit": null,
87
  "accelerate_version": "0.32.1",
 
100
  "prefill": {
101
  "memory": {
102
  "unit": "MB",
103
+ "max_ram": 958.521344,
104
  "max_global_vram": null,
105
  "max_process_vram": null,
106
  "max_reserved": null,
 
109
  "latency": {
110
  "unit": "s",
111
  "count": 15,
112
+ "total": 0.6492260389998705,
113
+ "mean": 0.043281735933324704,
114
+ "stdev": 0.0031079231381562398,
115
+ "p50": 0.0412588659999642,
116
+ "p90": 0.04721052060000375,
117
+ "p95": 0.04774936790000765,
118
+ "p99": 0.047909406379997105,
119
  "values": [
120
+ 0.047663633000013306,
121
+ 0.04653085199998941,
122
+ 0.04794941599999447,
123
+ 0.045464217000017015,
124
+ 0.04615094199999703,
125
+ 0.046105225999951926,
126
+ 0.04577523900002234,
127
+ 0.0412588659999642,
128
+ 0.04001042999999527,
129
+ 0.04014367899998206,
130
+ 0.0411057000000028,
131
+ 0.03982140599998729,
132
+ 0.04035695800001804,
133
+ 0.040724245999967934,
134
+ 0.04016522899996744
135
  ]
136
  },
137
  "throughput": {
138
  "unit": "tokens/s",
139
+ "value": 46.20886747890588
140
  },
141
  "energy": {
142
  "unit": "kWh",
143
+ "cpu": 1.600299941168891e-06,
144
+ "ram": 6.687847729458553e-08,
145
  "gpu": 0.0,
146
+ "total": 1.6671784184634767e-06
147
  },
148
  "efficiency": {
149
  "unit": "tokens/kWh",
150
+ "value": 1199631.6518080062
151
  }
152
  },
153
  "decode": {
154
  "memory": {
155
  "unit": "MB",
156
+ "max_ram": 958.521344,
157
  "max_global_vram": null,
158
  "max_process_vram": null,
159
  "max_reserved": null,
 
162
  "latency": {
163
  "unit": "s",
164
  "count": 15,
165
+ "total": 0.40746452299998737,
166
+ "mean": 0.02716430153333249,
167
+ "stdev": 0.0024496243438844713,
168
+ "p50": 0.02551993100001937,
169
+ "p90": 0.030082074999972973,
170
+ "p95": 0.0306673562999606,
171
+ "p99": 0.031544356059968096,
172
  "values": [
173
+ 0.030197534999956588,
174
+ 0.029908884999997554,
175
+ 0.028831417999981568,
176
+ 0.02918251500000224,
177
+ 0.028979173999971408,
178
+ 0.028825026000049547,
179
+ 0.03176360599996997,
180
+ 0.024657897999986744,
181
+ 0.02489800699999023,
182
+ 0.02551993100001937,
183
+ 0.024732778000043254,
184
+ 0.024578318000010313,
185
+ 0.025177860999974655,
186
+ 0.02530094100001179,
187
+ 0.024910630000022138
188
  ]
189
  },
190
  "throughput": {
191
  "unit": "tokens/s",
192
+ "value": 36.81302089703762
193
  },
194
  "energy": {
195
  "unit": "kWh",
196
+ "cpu": 9.547856118943962e-07,
197
+ "ram": 3.99040916730172e-08,
198
  "gpu": 0.0,
199
+ "total": 9.94689703567413e-07
200
  },
201
  "efficiency": {
202
  "unit": "tokens/kWh",
203
+ "value": 1005338.646226599
204
  }
205
  },
206
  "per_token": {
 
208
  "latency": {
209
  "unit": "s",
210
  "count": 15,
211
+ "total": 0.40153535700005705,
212
+ "mean": 0.026769023800003802,
213
+ "stdev": 0.002365268592983841,
214
+ "p50": 0.025195664000023044,
215
+ "p90": 0.029590286799998467,
216
+ "p95": 0.030177723599985027,
217
+ "p99": 0.031049157519963726,
218
  "values": [
219
+ 0.02971088399999644,
220
+ 0.029409391000001506,
221
+ 0.02836319199997206,
222
+ 0.02869021399999383,
223
+ 0.028483186000016758,
224
+ 0.028349066000032508,
225
+ 0.0312670159999584,
226
+ 0.02435206500001641,
227
+ 0.024594798999999057,
228
+ 0.025195664000023044,
229
+ 0.024424681000027704,
230
+ 0.024270462000004045,
231
+ 0.024850689000004422,
232
+ 0.024977255000010246,
233
+ 0.024596793000000616
234
  ]
235
  },
236
  "throughput": {
237
  "unit": "tokens/s",
238
+ "value": 37.35661066578969
239
  },
240
  "energy": null,
241
  "efficiency": null