Upload /Llama-2-13b-hf/int8_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json with huggingface_hub
Browse files
Llama-2-13b-hf/int8_batch_size_1_sq_len_256_new_tokens_256/benchmark_report.json
ADDED
@@ -0,0 +1,444 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"prefill": {
|
3 |
+
"memory": {
|
4 |
+
"unit": "MB",
|
5 |
+
"max_ram": 1041.276928,
|
6 |
+
"max_vram": 14522.7776,
|
7 |
+
"max_reserved": 14029.94688,
|
8 |
+
"max_allocated": 13954.652672
|
9 |
+
},
|
10 |
+
"latency": {
|
11 |
+
"unit": "s",
|
12 |
+
"mean": 0.07958155398898656,
|
13 |
+
"stdev": 0.0009138226668875471,
|
14 |
+
"values": [
|
15 |
+
0.08430230712890625,
|
16 |
+
0.08248524475097656,
|
17 |
+
0.08187494659423829,
|
18 |
+
0.08749874877929688,
|
19 |
+
0.07965388488769531,
|
20 |
+
0.07932313537597656,
|
21 |
+
0.07968665313720703,
|
22 |
+
0.07956275177001954,
|
23 |
+
0.07960169219970703,
|
24 |
+
0.07977471923828125,
|
25 |
+
0.07971942138671875,
|
26 |
+
0.07953305816650391,
|
27 |
+
0.07949005126953125,
|
28 |
+
0.07920845031738281,
|
29 |
+
0.07940608215332032,
|
30 |
+
0.07938047790527344,
|
31 |
+
0.07988735961914062,
|
32 |
+
0.07939788818359375,
|
33 |
+
0.0798054428100586,
|
34 |
+
0.0796600341796875,
|
35 |
+
0.07961609649658204,
|
36 |
+
0.07959059143066406,
|
37 |
+
0.07961190032958984,
|
38 |
+
0.0795832290649414,
|
39 |
+
0.07932825469970703,
|
40 |
+
0.07925965118408203,
|
41 |
+
0.07929138946533203,
|
42 |
+
0.07948902130126953,
|
43 |
+
0.07947264099121094,
|
44 |
+
0.07950233459472657,
|
45 |
+
0.07943679809570313,
|
46 |
+
0.07938662719726562,
|
47 |
+
0.07951974487304687,
|
48 |
+
0.07955865478515625,
|
49 |
+
0.0796049575805664,
|
50 |
+
0.07994572448730469,
|
51 |
+
0.07951769256591797,
|
52 |
+
0.0794419174194336,
|
53 |
+
0.07928627014160156,
|
54 |
+
0.07946240234375,
|
55 |
+
0.07945420837402344,
|
56 |
+
0.07990067291259766,
|
57 |
+
0.07939071655273437,
|
58 |
+
0.07948492431640625,
|
59 |
+
0.07947980499267578,
|
60 |
+
0.07966719818115234,
|
61 |
+
0.07946463775634766,
|
62 |
+
0.07955455780029297,
|
63 |
+
0.07940505981445313,
|
64 |
+
0.07941120147705077,
|
65 |
+
0.07944499206542968,
|
66 |
+
0.07950540924072266,
|
67 |
+
0.07932825469970703,
|
68 |
+
0.07941120147705077,
|
69 |
+
0.07943577575683594,
|
70 |
+
0.07934054565429688,
|
71 |
+
0.07943270111083985,
|
72 |
+
0.07930368041992188,
|
73 |
+
0.07940300750732422,
|
74 |
+
0.07925555419921874,
|
75 |
+
0.07984333038330078,
|
76 |
+
0.07948902130126953,
|
77 |
+
0.0793364486694336,
|
78 |
+
0.079329345703125,
|
79 |
+
0.0793855972290039,
|
80 |
+
0.079425537109375,
|
81 |
+
0.07955353546142578,
|
82 |
+
0.07921151733398438,
|
83 |
+
0.07921663665771485,
|
84 |
+
0.07935590362548828,
|
85 |
+
0.07949517059326172,
|
86 |
+
0.07944806671142578,
|
87 |
+
0.07896063995361328,
|
88 |
+
0.07923814392089844,
|
89 |
+
0.07918080139160157,
|
90 |
+
0.07926681518554687,
|
91 |
+
0.0792248306274414,
|
92 |
+
0.07914086151123047,
|
93 |
+
0.07947468566894532,
|
94 |
+
0.0792074203491211,
|
95 |
+
0.0794224624633789,
|
96 |
+
0.07924018859863281,
|
97 |
+
0.07938976287841797,
|
98 |
+
0.07939993286132813,
|
99 |
+
0.07939071655273437,
|
100 |
+
0.07945523071289062,
|
101 |
+
0.0798545913696289,
|
102 |
+
0.07955455780029297,
|
103 |
+
0.07916134643554687,
|
104 |
+
0.0793169937133789,
|
105 |
+
0.07929344177246093,
|
106 |
+
0.07943065643310547,
|
107 |
+
0.07970214080810546,
|
108 |
+
0.07957504272460937,
|
109 |
+
0.0793364486694336,
|
110 |
+
0.07951360321044922,
|
111 |
+
0.07940198516845703,
|
112 |
+
0.07918694305419922,
|
113 |
+
0.07916031646728515,
|
114 |
+
0.07951667022705078,
|
115 |
+
0.07941529846191406,
|
116 |
+
0.07922380828857421,
|
117 |
+
0.07938355255126953,
|
118 |
+
0.07930879974365235,
|
119 |
+
0.07947673797607421,
|
120 |
+
0.07936316680908204,
|
121 |
+
0.07918489837646485,
|
122 |
+
0.07923916625976563,
|
123 |
+
0.07955661010742188,
|
124 |
+
0.07933849334716797,
|
125 |
+
0.07938764953613281,
|
126 |
+
0.0798760986328125,
|
127 |
+
0.07915827178955079,
|
128 |
+
0.07928934478759765,
|
129 |
+
0.07933235168457031,
|
130 |
+
0.07937229156494141,
|
131 |
+
0.07932313537597656,
|
132 |
+
0.07978291320800782,
|
133 |
+
0.07941840362548828,
|
134 |
+
0.0791910400390625,
|
135 |
+
0.07937638092041016,
|
136 |
+
0.07940402984619141,
|
137 |
+
0.07953715515136718,
|
138 |
+
0.07942655944824219,
|
139 |
+
0.07941426849365234,
|
140 |
+
0.0791756820678711
|
141 |
+
]
|
142 |
+
},
|
143 |
+
"throughput": {
|
144 |
+
"unit": "tokens/s",
|
145 |
+
"value": 3216.825849309607
|
146 |
+
},
|
147 |
+
"energy": null,
|
148 |
+
"efficiency": null
|
149 |
+
},
|
150 |
+
"decode": {
|
151 |
+
"memory": {
|
152 |
+
"unit": "MB",
|
153 |
+
"max_ram": 1041.276928,
|
154 |
+
"max_vram": 15311.306752,
|
155 |
+
"max_reserved": 14818.476032,
|
156 |
+
"max_allocated": 14774.906368
|
157 |
+
},
|
158 |
+
"latency": {
|
159 |
+
"unit": "s",
|
160 |
+
"mean": 20.578271308898923,
|
161 |
+
"stdev": 0,
|
162 |
+
"values": [
|
163 |
+
20.578271308898923
|
164 |
+
]
|
165 |
+
},
|
166 |
+
"throughput": {
|
167 |
+
"unit": "tokens/s",
|
168 |
+
"value": 12.39171144029611
|
169 |
+
},
|
170 |
+
"energy": null,
|
171 |
+
"efficiency": null
|
172 |
+
},
|
173 |
+
"per_token": {
|
174 |
+
"memory": null,
|
175 |
+
"latency": {
|
176 |
+
"unit": "s",
|
177 |
+
"mean": 0.08069910317215263,
|
178 |
+
"stdev": 0.0031678996097352454,
|
179 |
+
"values": [
|
180 |
+
0.08110387420654297,
|
181 |
+
0.08053145599365234,
|
182 |
+
0.07975628662109376,
|
183 |
+
0.07939788818359375,
|
184 |
+
0.07976652526855468,
|
185 |
+
0.0795832290649414,
|
186 |
+
0.07942758178710937,
|
187 |
+
0.07965491485595703,
|
188 |
+
0.0797122573852539,
|
189 |
+
0.0800512924194336,
|
190 |
+
0.07998249816894532,
|
191 |
+
0.07985049438476563,
|
192 |
+
0.0795863037109375,
|
193 |
+
0.08005734252929687,
|
194 |
+
0.07990169525146484,
|
195 |
+
0.08020172882080077,
|
196 |
+
0.07986994934082031,
|
197 |
+
0.07966617584228515,
|
198 |
+
0.07967231750488281,
|
199 |
+
0.07970611572265625,
|
200 |
+
0.07965081787109375,
|
201 |
+
0.07981056213378906,
|
202 |
+
0.07977677154541016,
|
203 |
+
0.07957810974121093,
|
204 |
+
0.0795699234008789,
|
205 |
+
0.07948595428466797,
|
206 |
+
0.07930060577392578,
|
207 |
+
0.07952178955078125,
|
208 |
+
0.07954431915283203,
|
209 |
+
0.07955174255371093,
|
210 |
+
0.079884033203125,
|
211 |
+
0.07957433319091797,
|
212 |
+
0.07975596618652343,
|
213 |
+
0.07943679809570313,
|
214 |
+
0.07976140594482421,
|
215 |
+
0.08015270233154297,
|
216 |
+
0.08003571319580079,
|
217 |
+
0.07970201873779297,
|
218 |
+
0.0798054428100586,
|
219 |
+
0.07931801605224609,
|
220 |
+
0.07967436981201172,
|
221 |
+
0.08002457427978515,
|
222 |
+
0.07965081787109375,
|
223 |
+
0.0796753921508789,
|
224 |
+
0.0796385269165039,
|
225 |
+
0.07966719818115234,
|
226 |
+
0.07971942138671875,
|
227 |
+
0.07963442993164062,
|
228 |
+
0.07957810974121093,
|
229 |
+
0.0796756820678711,
|
230 |
+
0.08157154846191406,
|
231 |
+
0.07943065643310547,
|
232 |
+
0.0796211166381836,
|
233 |
+
0.07949619293212891,
|
234 |
+
0.0795832290649414,
|
235 |
+
0.07955046081542969,
|
236 |
+
0.07961702728271484,
|
237 |
+
0.0793007049560547,
|
238 |
+
0.07954217529296875,
|
239 |
+
0.07962425231933594,
|
240 |
+
0.07992006683349609,
|
241 |
+
0.07945011138916015,
|
242 |
+
0.07961190032958984,
|
243 |
+
0.07950847625732421,
|
244 |
+
0.07968460845947266,
|
245 |
+
0.07946649932861329,
|
246 |
+
0.0799139862060547,
|
247 |
+
0.07935794830322265,
|
248 |
+
0.07949209594726563,
|
249 |
+
0.07939891052246094,
|
250 |
+
0.07950847625732421,
|
251 |
+
0.07952998352050782,
|
252 |
+
0.07955046081542969,
|
253 |
+
0.07965286254882813,
|
254 |
+
0.07960985565185547,
|
255 |
+
0.07947468566894532,
|
256 |
+
0.079840576171875,
|
257 |
+
0.07972431945800781,
|
258 |
+
0.07975926208496094,
|
259 |
+
0.07943577575683594,
|
260 |
+
0.07953305816650391,
|
261 |
+
0.07957094573974609,
|
262 |
+
0.0794972152709961,
|
263 |
+
0.07967040252685546,
|
264 |
+
0.0797080307006836,
|
265 |
+
0.08005017852783203,
|
266 |
+
0.0800348129272461,
|
267 |
+
0.07955967712402344,
|
268 |
+
0.07972761535644532,
|
269 |
+
0.07975654602050782,
|
270 |
+
0.0799208984375,
|
271 |
+
0.08012799835205078,
|
272 |
+
0.07978598022460938,
|
273 |
+
0.07999590301513672,
|
274 |
+
0.08003993225097657,
|
275 |
+
0.07987302398681641,
|
276 |
+
0.0799805450439453,
|
277 |
+
0.0796610565185547,
|
278 |
+
0.08536166381835937,
|
279 |
+
0.09626316833496094,
|
280 |
+
0.09969971466064453,
|
281 |
+
0.0990904312133789,
|
282 |
+
0.0937154541015625,
|
283 |
+
0.09457049560546875,
|
284 |
+
0.09063629150390624,
|
285 |
+
0.08748646545410156,
|
286 |
+
0.08769945526123046,
|
287 |
+
0.08765952301025391,
|
288 |
+
0.08879206085205078,
|
289 |
+
0.0882165756225586,
|
290 |
+
0.08778546905517579,
|
291 |
+
0.08755712127685547,
|
292 |
+
0.08753971099853515,
|
293 |
+
0.08756752014160156,
|
294 |
+
0.08813654327392578,
|
295 |
+
0.08765542602539063,
|
296 |
+
0.0876575698852539,
|
297 |
+
0.08772291564941406,
|
298 |
+
0.0878704605102539,
|
299 |
+
0.08762060546875,
|
300 |
+
0.07979007720947266,
|
301 |
+
0.08000102233886719,
|
302 |
+
0.07983103942871093,
|
303 |
+
0.07972147369384766,
|
304 |
+
0.07997030639648438,
|
305 |
+
0.08001023864746094,
|
306 |
+
0.07986176300048828,
|
307 |
+
0.07980057525634765,
|
308 |
+
0.07989119720458984,
|
309 |
+
0.07999180603027344,
|
310 |
+
0.07978598022460938,
|
311 |
+
0.07981491088867188,
|
312 |
+
0.0795758056640625,
|
313 |
+
0.08022118377685547,
|
314 |
+
0.07964876556396484,
|
315 |
+
0.07961497497558594,
|
316 |
+
0.07991500854492188,
|
317 |
+
0.07960591888427734,
|
318 |
+
0.07980630493164062,
|
319 |
+
0.08019353485107422,
|
320 |
+
0.07978189086914063,
|
321 |
+
0.07971635437011719,
|
322 |
+
0.07963238525390624,
|
323 |
+
0.080036865234375,
|
324 |
+
0.07983718109130859,
|
325 |
+
0.08006553649902344,
|
326 |
+
0.0795688934326172,
|
327 |
+
0.07983001708984375,
|
328 |
+
0.0798883819580078,
|
329 |
+
0.07988428497314454,
|
330 |
+
0.07978495788574219,
|
331 |
+
0.09140019226074218,
|
332 |
+
0.08715980529785156,
|
333 |
+
0.08014335632324218,
|
334 |
+
0.08041471862792969,
|
335 |
+
0.07968386840820313,
|
336 |
+
0.08000994873046875,
|
337 |
+
0.07981881713867188,
|
338 |
+
0.08014125061035156,
|
339 |
+
0.07968972778320313,
|
340 |
+
0.07954637145996094,
|
341 |
+
0.08014335632324218,
|
342 |
+
0.07965081787109375,
|
343 |
+
0.07965900421142579,
|
344 |
+
0.08039014434814454,
|
345 |
+
0.07993654632568359,
|
346 |
+
0.07986685180664063,
|
347 |
+
0.07960166168212891,
|
348 |
+
0.07964569854736328,
|
349 |
+
0.0799459228515625,
|
350 |
+
0.07970899200439453,
|
351 |
+
0.07979724884033203,
|
352 |
+
0.07979519653320312,
|
353 |
+
0.07956582641601563,
|
354 |
+
0.07981369781494141,
|
355 |
+
0.07977903747558594,
|
356 |
+
0.0797334747314453,
|
357 |
+
0.07958528137207031,
|
358 |
+
0.07964672088623047,
|
359 |
+
0.07968265533447266,
|
360 |
+
0.0796589126586914,
|
361 |
+
0.07970918273925781,
|
362 |
+
0.07967334747314453,
|
363 |
+
0.0798392333984375,
|
364 |
+
0.07983001708984375,
|
365 |
+
0.07970428466796875,
|
366 |
+
0.07966083526611328,
|
367 |
+
0.07978495788574219,
|
368 |
+
0.07953202819824219,
|
369 |
+
0.08008704376220703,
|
370 |
+
0.07938047790527344,
|
371 |
+
0.07955455780029297,
|
372 |
+
0.07946956634521485,
|
373 |
+
0.07955379486083984,
|
374 |
+
0.07968844604492188,
|
375 |
+
0.07999078369140625,
|
376 |
+
0.07965388488769531,
|
377 |
+
0.07974617767333984,
|
378 |
+
0.0796148452758789,
|
379 |
+
0.07982899475097656,
|
380 |
+
0.07955168151855468,
|
381 |
+
0.07946323394775391,
|
382 |
+
0.07949823760986328,
|
383 |
+
0.07954329681396484,
|
384 |
+
0.07946546936035156,
|
385 |
+
0.07962009429931641,
|
386 |
+
0.07972249603271485,
|
387 |
+
0.07970435333251953,
|
388 |
+
0.07989014434814454,
|
389 |
+
0.08013209533691407,
|
390 |
+
0.07964569854736328,
|
391 |
+
0.07957504272460937,
|
392 |
+
0.07996518707275391,
|
393 |
+
0.0799805450439453,
|
394 |
+
0.07977369689941406,
|
395 |
+
0.08023961639404296,
|
396 |
+
0.07981362915039063,
|
397 |
+
0.07973990631103516,
|
398 |
+
0.07953612518310547,
|
399 |
+
0.07960269165039062,
|
400 |
+
0.07953510284423829,
|
401 |
+
0.07946956634521485,
|
402 |
+
0.0797655029296875,
|
403 |
+
0.07961907196044922,
|
404 |
+
0.07984947204589844,
|
405 |
+
0.07963648223876953,
|
406 |
+
0.07960166168212891,
|
407 |
+
0.07986688232421875,
|
408 |
+
0.07964057922363281,
|
409 |
+
0.07991913604736328,
|
410 |
+
0.079795166015625,
|
411 |
+
0.07962841796875,
|
412 |
+
0.07947968292236328,
|
413 |
+
0.07964979553222656,
|
414 |
+
0.080321533203125,
|
415 |
+
0.07959574127197265,
|
416 |
+
0.07955945587158203,
|
417 |
+
0.07946982574462891,
|
418 |
+
0.07974169921875,
|
419 |
+
0.07978905487060547,
|
420 |
+
0.0801976318359375,
|
421 |
+
0.07961702728271484,
|
422 |
+
0.07957094573974609,
|
423 |
+
0.07993958282470703,
|
424 |
+
0.07983411407470703,
|
425 |
+
0.07973900604248046,
|
426 |
+
0.07973670196533203,
|
427 |
+
0.07975833892822265,
|
428 |
+
0.07979827117919921,
|
429 |
+
0.07975628662109376,
|
430 |
+
0.0798895034790039,
|
431 |
+
0.07979920196533204,
|
432 |
+
0.07995420837402344,
|
433 |
+
0.07978262329101563,
|
434 |
+
0.0798424301147461
|
435 |
+
]
|
436 |
+
},
|
437 |
+
"throughput": {
|
438 |
+
"unit": "tokens/s",
|
439 |
+
"value": 12.391711440296111
|
440 |
+
},
|
441 |
+
"energy": null,
|
442 |
+
"efficiency": null
|
443 |
+
}
|
444 |
+
}
|